TRF1: It Was the Best of Time(s)…

In: Timing & Time Perception
View More View Less
  • 1 INSERM, France
  • | 2 Laboratoire des Neurosciences Cognitives (UMR 7291), Aix-Marseille University& CNRS, Marseille, France
Open Access

The Timing Research Forum (TRF; is an international (Fig. 1), gender-inclusive, and open academic society for timing research, founded in 2016 by Argiro Vatakis and Sundeep Teki (Teki, 2016). Following a call to its Committee members in May 2016, we agreed to host the 1st Conference of the Timing Research Forum in Strasbourg, France (TRF1; First on the agenda was to decide on eminent keynote speakers to lend credibility to this very first TRF conference. We wanted one talk from the field of psychology and the other from the neurosciences, and so were delighted that both Lera Boroditsky and Warren Meck accepted our invitations immediately. Lera kicked off the conference for us, with an extremely entertaining talk about the spatial representation of time in different societies and cultures and how the linguistic metaphors we use to describe time influence our conception of time. Warren highlighted the key role of the striatal dopaminergic system for timing, illustrating his talk with data from an impressive variety of methodological techniques from the clinical level (performance in patients with Parkinson’s Disease) right down to the cellular (optogenetic studies in mice). Coincidentally, the role of dopamine in timing was also the subject of our third keynote talk. One of the mission statements of TRF is to promote the work of young researchers, and Argiro and Sundeep had the great idea to invite an early-career researcher to give a keynote talk. The TRF committee members were invited to submit recommendations of key timing papers published in 2016, and Sofia Soares was the unanimous choice to present her PhD work with Joseph Paton on the timing functions of midbrain dopamine neurons, published in Science (Soares et al., 2016). She delivered an incredibly assured talk, demonstrating that the activity of dopaminergic neurons in the substantia nigra, though not of those in the ventral tegmental area, causally reflects and controls time-based judgments.

Figure 1.
Figure 1.

The TRF membership across the world as of October 2017.

Citation: Timing & Time Perception 6, 3-4 (2018) ; 10.1163/22134468-00603001

We knew TRF had a large membership with more than 650 members and that the timing community was growing. But for a specialized conference, we judged three days of single-track sessions would be sufficient. We therefore planned and announced the dates (October 23–25, 2017) accordingly. However, the incredible response to our call for abstracts (>200 submissions) meant that we had far more abstracts than time available. We therefore had to make the difficult decision to program parallel sessions. Some tough scheduling choices would have to be made! We were particularly overwhelmed by the excellent quality of submissions for symposia and wish we could have accepted them all. In the end, we chose a balanced selection of neuroscientific, psychological and computational symposia that spanned timing from the milliseconds range right up to circadian rhythms (see Fig. 2). The 15 symposia (encompassing 52 individual talks, 35% female presenters) covered a wide spectrum of timing phenomena: how duration is processed (e.g., spatial, motor or embodied representations of time; scalar properties of time) and how it can be used for learning or making predictions, how the order of events in time impacts our conscious experience (e.g., intentional binding; the continuous nature of consciousness), and how sensory and neural rhythms guide our sense of time in both humans and animals. Actually, very similar themes were reflected in the submissions for short oral presentations. And once the selected abstracts (36 talks, 42% female presenters) had been organized into seven themed sessions, the high quality of the presentations made each session actually feel like a premeditated symposium!

Figure 2.
Figure 2.

Frequency of time scales associated with TRF1 abstract submissions.

Citation: Timing & Time Perception 6, 3-4 (2018) ; 10.1163/22134468-00603001

We also had a large number of posters (120 posters, 53% female presenters), which were presented in the beautiful main hall of Strasbourg’s UNESCO heritage University Palace over two days. The hall was also the location for the lunch breaks, meaning researchers could gather together and discuss the displayed posters outside of the designated poster sessions. Posters covered an even more diverse range of topics than the talks or symposia. On the first day, we had posters describing how the perception of duration can be modulated by a variety of distinct factors (memory, emotion, motivation, meditation, cognitive effort, salience, visual characteristics, magnitude) and how it is impaired in a number of psychiatric and neurological disorders (e.g., focal lesions, dementia, Parkinson’s Disease, schizophrenia, bipolar disorder, depression, ADHD, autism). On the second day, we had neuroscientific investigations of duration processing encompassing several different methodological techniques (electrophysiology, neuroimaging, lesions, pharmacology) and a large number of posters on rhythm processing and temporal predictions. This session also included posters on computational models of timing and the role of time perception in computer science and robotics. Finally, we had several posters on multisensory timing, the processing of order and intentional binding. The variety of topics meant there was something for everyone (see Fig. 3), as well as providing the opportunity to discover new ways in which timing can be seen to play a fundamental role in behaviour. Eleven highly rated posters were selected for the poster blitz session, in which junior researchers gave a five-minute talk to publicise their work.

Figure 3.
Figure 3.

A word cloud featuring keywords from all abstracts presented at TRF1 where the size of a keyword is proportional to its frequency of occurrence across all submissions.

Citation: Timing & Time Perception 6, 3-4 (2018) ; 10.1163/22134468-00603001

TRF fosters an open research community, evidenced by their collaboration with the Open Science Foundation to host electronic versions of the posters and talks presented at the conference ( To date, more than 100 submissions have been shared openly with the wider scientific community, demonstrating the commitment of TRF and its members to open science and open access. Additionally, TRF has built a very active online social media presence on Twitter, Facebook, and ResearchGate (more than 1000 followers) and the power of these networks was leveraged before and during the conference to engage with the participants and the wider timing research community. A series of Q&A interview columns featuring distinguished speakers at the TRF1 conference has been openly shared, highlighting the thoughts and opinions of leading scientists on the current state of timing research as well as advice for early-career researchers.

We would like to finish by thanking Argiro and Sundeep, gratefully acknowledging their generosity and dedication in setting up TRF from scratch. Without them, we would not have had the platform that allowed us to bring together 270 people from 28 different countries and five continents (Fig. 4), to describe such diverse aspects of timing research. Thanks to both Sundeep and Argiro for providing a virtual home for the timing research community. But even more importantly, we would like to thank all of the participants for their support and enthusiasm for TRF1. The impressive quality and quantity of the presentations (see this issue), and the stimulating and positive atmosphere during the conference itself, exceeded all our expectations. The huge success of this very first TRF conference attests to the undeniable need for a regular meeting for our international timing community.

Figure 4.
Figure 4.

TRF1 participants came to Strasbourg from across the globe. The vast majority (~70%) were from Europe with, unsurprisingly, the largest proportion (30%) being from the host country France. Other well-represented countries included the UK (11%), Germany (8%), USA (7%), Italy (6%), the Netherlands (5%), Japan (5%) and Brazil (4%).

Citation: Timing & Time Perception 6, 3-4 (2018) ; 10.1163/22134468-00603001


Soares, S., Atallah, B. V., & Paton, J. J. (2016). Midbrain dopamine neurons control judgment of time. Science, 354, 1273–1277.

Teki, S. (2016). A citation-based analysis and review of significant papers on timing and time perception. Front Neurosci., 10, 330. doi: 10.3389/fnins.2016.00330.


(in order of presentation)

Neural Entrainment as a Mechanism of Efficient Stimulus Processing

Benedikt Zoefel 1 , Alan Archer-Boyd 1 , Matthew H. Davis 1 , Monica N. O’Connell 2 , Annamaria Barczak 2 , Tammy McGinnis 2 , Deborah Ross 2 , Peter Lakatos 2 , Molly J. Henry 3 and Sanne Ten Oever 4

1MRC Cognition and Brain Sciences Unit, Cambridge, UK

2Nathan Kline Institute for Psychiatric Research, Orangeburg, NY, USA

3Brain and Mind Institute, University of Western Ontario, Canada

4Department of Cognitive Neuroscience, Faculty of Psychology and Neuroscience, Maastricht University, The Netherlands

Symposium organizer: Benedikt Zoefel

Neural entrainment, the alignment between neural oscillations and rhythmic stimulus input, is often assigned a critical role for stimulus processing, selection, and predictions: By aligning the high-excitability oscillatory phase with the timing of expected events, important stimulus input can be selectively amplified (e.g., Schroeder & Lakatos, 2009). Nevertheless, studies are often criticized, due to an apparent failure to distinguish neural entrainment from other processes that can potentially produce similar data, or due to a lack of understanding of the underlying neural mechanisms. In this symposium, we use state-of-the-art methods to address these issues in several studies. In the first talk (O’Connell), we demonstrate in intracranial non-human primate recordings that a network comprising non-specific thalamic and cortical regions is critically involved in the phase-reset of neural oscillations, a fundamental mechanism for an efficient adjustment to stimulus input and an often hypothesized process underlying neural entrainment (e.g., Lakatos et al., 2009). In two subsequent talks, we show that neural entrainment is not merely a superposition of regular evoked responses, an issue raised previously (e.g., Capilla et al., 2011): In electroencephalographic (EEG) data (Henry), we show that neural entrainment can be dissociated from evoked responses by independently manipulating the perceived beat and acoustic properties of a rhythmic musical stimulus. In magnetoencephalographic (MEG) recordings (Ten Oever), we show that neural entrainment persists when participants do not consciously perceive the entraining stimulus (which strongly reduces evoked neural activity). In the last talk of the symposium (Zoefel), we use combined transcranial alternating current stimulation (tACS) and functional magnetic resonance imaging (fMRI) recordings to demonstrate that neural entrainment is not only an epi-phenomenon but causally involved in speech processing: Manipulating the phase relation between neural oscillations and speech rhythm affects the blood-oxygen-level dependent (BOLD) response to intelligible (but not unintelligible) speech. Together, our symposium provides important and complementary evidence that neural entrainment is more than a simple repetition of evoked responses: Instead, it reflects an efficient mechanism for the processing of rhythmic stimulus input, including speech sounds, and involving thalamocortical connections as one of the underlying neural circuitries.


Rhythms, synchrony, electrophysiology, adult, non-human primates, 100s of ms-secs

1.Mapping the Circuitry of Oscillatory Phase Reset (M. N. O’Connell, A. Barczak, T. McGinnis, D. Ross and P. Lakatos)

To be able to align internal to external rhythms, the brain utilizes a mechanism called phase reset to modulate the phase of neuronal oscillations. It is speculated that inputs originating in the non-specific thalamus are responsible for this mechanism. We tested this hypothesis by recording A1 and MGN neuroelectric activity simultaneously, in awake macaques who were presented with a rhythmic stream of LED flashes. We found that some single neurons in MGN responded to LED flashes, but only on a subset of trials. During trials where the visually-responsive MGN neurons fired, concurrently recorded A1 ensemble activity showed signatures of phase reset, accompanied by significant phase alignment to stimuli. Subsequent to LED related firing in the MGN, a subpopulation of A1 units also fired. Firing of both MGN and A1 units was accompanied by stimulus locked high-gamma bursts, indicating likely relevance for the mechanism of phase reset.

2.Separating Stimulus-Driven and Entrained Neural Responses Using Musical Rhythms (M. J. Henry)

Interest in the role of neural entrainment in perception of rhythmic environmental stimuli is on the rise. However, the nature of the neural signals observed in response to rhythmic stimuli is still a matter of debate. Specifically, it is unclear the extent to which these neural responses reflect entrainment of neural oscillations, or may instead reflect superposition of transient responses evoked by stimulus onsets. I will present electroencephalography (EEG) data collected during listening to musical rhythms, where we independently manipulated the strength of the perceived beat and acoustic properties of the rhythms in order to systematically change the energy present in the spectrum of the stimulus envelope at beat- related frequencies. We quantified the contributions of stimulus energy and perceived beat strength to the neural signal. Our results suggest that neural responses to musical rhythms reflect more than a superposition of transient evoked responses, and thus indicate entrainment of neural oscillations.

3.Temporal Expectations Influence Entrainment Presence and Strength (S. Ten Oever)

Neuronal oscillations reflect fluctuations in membrane potentials between higher and lower excitable states. It has been proposed that high excitable phases of ongoing oscillations align to rhythmic input streams to optimize sensory processing, so-called entrainment. With MEG recordings we show that entrainment optimizes perception for rhythmic streams that are still below perceptual detection thresholds. Interestingly, it was primarily phase alignment, as measured with inter-trial coherence, and not power, that increased prior to detection. These results suggest that the measured entrainment was not solely a consequence of sensory evoked responses and that pre-detection top-down temporal expectations can induce this entrainment. In an EEG study we corroborate on the top-down influences on entrainment by identifying the influence of temporal expectations on entrainment strength. Both studies point to a relevant role of entrainment for behavior and not a mere byproduct of the responses to the sensory input.

4.Phase Entrainment of Neural Oscillations Is Causally Relevant for Neural Responses to Intelligible Speech (B. Zoefel, A. Archer-Boyd and M. H. Davis)

The entrainment of neural oscillations is often considered critical for speech processing. Nevertheless, only if we manipulate entrainment as a dependent variable and observe consequences for speech processing, can we can conclude that there is a causal relation between the two. We manipulated entrainment using transcranial alternating current stimulation (tACS) and systematically varied the phase relation between tACS and an intelligible or unintelligible rhythmic speech stimulus. Recording fMRI data simultaneously, we tested whether this manipulation of entrainment affects neural activity (reflected in the BOLD response). We found that, for intelligible speech, the relation between tACS phase and speech rhythm significantly modulates the magnitude of the BOLD response in the Superior Temporal Gyrus. Importantly, a significant interaction showed that the effect was reduced and absent for unintelligible speech and during sham stimulation. Our results therefore suggest that neural entrainment has a specific, causal influence on neural responses to intelligible speech.


Capilla, A., Pazo-Alvarez, P., Darriba, A., Campo, P., & Gross, J. (2011). Steady-state visual evoked potentials can be explained by temporal superposition of transient event-related responses. PloS One, 6, e14543.

Lakatos, P., O’Connell, M. N., Barczak, A., Mills, A., Javitt, D. C., & Schroeder, C. E. (2009). The leading sense: supramodal control of neurophysiological context by attention. Neuron, 64, 419–430.

Schroeder, C. E., & Lakatos, P. (2009). Low-frequency neuronal oscillations as instruments of sensory selection. Trends in Neurosciences, 32, 9–18.

Interrelations Between the Representation of Time and Space

Martin Riemer 1 , Esther Kühn 1 , Jonathan Shine 1 , Thomas Wolbers 1,2 , Roberto Bottini 3 , Baptiste Gauthier 4,5 , Karin Pestke 4 and Virginie van Wassenhove 4

1Aging & Cognition Research Group, German Center for Neurodegenerative Diseases (DZNE), Magdeburg, Germany

2Center for Behavioral Brain Sciences, Magdeburg, Germany

3Center for Mind/Brain Sciences, University of Trento, Italy

4Cognitive Neuroimaging Unit, CEA DRF/I2BM, INSERM, Université Paris-Sud, Université Paris-Saclay, NeuroSpin center, F-91191 Gif/Yvette, France

5Laboratory of Cognitive Neuroscience, Faculty of Life Science, Brain Mind Institute, Ecole Polytechnique Federale de Lausanne, Geneva, Switzerland

Symposium organizer: Martin Riemer

The concepts of space and time are highly intertwined in the human mind. We all are familiar with graphical illustrations of temporal sequences, in which successive events are ordered next to each other along a spatial continuum. The localization of events in time (before/after) and the magnitude of durations (short/long) are often coded with a left-to-right spatial order on a ‘mental time line’ (Bonato et al., 2012). Linguistic approaches point out that we often use spatial metaphors to describe relations in time. The future lies in front of us and past events might be far away. Spatial metaphors for time are accompanied by body gestures, e.g., pointing backwards to indicate that something happened in the past. All these examples demonstrate the general tendency of the human mind to conceptualize in space what is perceived in time (Casasanto & Boroditsky, 2008). In support of this view, neuroimaging studies have discovered neural populations in the parietal cortex that code both for temporal and spatial magnitudes, and damages in these areas can impair time perception as well as spatial cognition. More recently, the discovery of place and grid cells in the hippocampus and the entorhinal cortex, and the observation that these cells exhibit spatial and temporal tuning curves provide new insights to the neuronal substrates of interrelations between time and space (Kraus et al., 2015). Yet, this wealth of empirical observations is still lacking a coherent theoretical framework that could explain (i) universal and culture-specific aspects of space-time interactions, (ii) whether the cognitive systems for time and space interact in a symmetric or an asymmetric manner, (iii) and whether different forms of space-time interactions (from magnitude judgments to mental time travel) can be traced back to one core neural system or are based on largely different mechanisms. In our symposium, we will present and discuss the results from recent fMRI and MEG/EEG studies, directly comparing neuroanatomical contributions to time and space representations in healthy participants (Riemer). As vision is a pertinent factor for the formation and maintenance of spatial representations, the impact of visual experience of space for time-space interactions will be addressed by comparing sighted and visually impaired participants (Bottini). Interrelations between time and space are not confined to direct perception. They also affect our ability to imagine and mentally represent these dimensions. The second part of our symposium will therefore focus on the question of how mental travel through time is influenced by spatial factors and vice versa. Again, we will discuss the ability for mental travel through time and space in healthy participants (Gauthier). Evaluation of competing hypotheses can produce a common working framework to investigate the links between time and space that are fundamental aspects of human cognition.


Duration, behavior, MRI, blind patients, neglect patients, secs-mins

1.Integration Processes of Travel Time and Traveled Distance (M. Riemer, E. Kühn, J. Shine and T. Wolbers)

The compelling interdependency between the perception of time and space becomes especially evident in the domain of spatial navigation, because moving in space requires time. Knowledge about the length of a covered distance contains valuable information about the corresponding travel time and vice versa. We used fMRI to identify the overlapping and diverging neuronal networks underlying the processing of traveled distance and travel time. In a virtual environment, participants walked along a straight path, while their attention was directed either to travel time or to the distance covered. The results demonstrate lateralization differences in the prefrontal cortex (PFC) for temporal and spatial processing. While attention to travel time is associated with bilateral activation, processing of traveled distance recruits only the right PFC. The presented study extends the research field to the integration of travel time and traveled distance in a navigational context.

2.The Role of Vision in Space-Time Interactions (R. Bottini)

Space and time are highly intertwined in the human mind. For instance, things that are longer in space also appear to last longer, and temporal succession is represented along Mental Time Lines that run from left to right or vice versa. It is still unclear how these Spatial-Temporal Associations (STAs) are established, and what is the role of experience in their development. For instance, does the way space and time interact in our mind change across sensory modalities (vision and audition)? Is visual experience necessary to establish space-time conceptual mappings? The study of visually deprived individuals may shed light on these issues. I will present behavioral and psychophysical experiments with sighted and early blind individuals suggesting that, although vision is not strictly necessary to develop STAs, space-time mappings may change as a function of the dominant modality in which we experience the world.

3.Ordering Events in Time and Space in Mental Travels (B. Gauthier, K. Pestke and V. van Wassenhove)

How the human brain represents time and space is essential to understand the conscious mind. When moving, mapping the environment yields a topological relationship between the traveled distance and the order of events. Nonetheless, in the absence of movement, e.g., while thinking

about future plans, the ordering of mental events may dissociate from their spatial dimension. In this M/EEG neuroimaging study, participants imagined themselves at different times and places (self-projection) and ordered memorized historical events from their mental standpoint. Early parametric changes of evoked responses amplitude, localized in medial temporal region, reflected past-to-future events succession while late spatial ordering evoked components were localized in parietal and frontal cortices. Overall, we report dedicated cortical signatures for the representation of conscious spatial and temporal distance and ordinality in the human brain. Crucially, the directionality of the psychological time arrow relies on neural mechanisms that are fundamentally dissociable from spatial directionalities.


Bonato, M., Zorzi, M., & Umilta, C. (2012). When time is space: Evidence for a mental time line. Neurosci. Biobehav. Rev., 36, 2257–2273.

Casasanto, D., & Boroditsky, L. (2008). Time in the mind: Using space to think about time. Cognition, 106, 579–593.

Kraus, B. J., Brandon, M. P., Robinson, R. J., 2nd, Connerney, M. A., Hasselmo, M. E., & Eichenbaum, H. (2015). During running in place, grid cells integrate elapsed time and distance run. Neuron, 88, 578–589.

Musical Rhythm: Evolutionary and Cross-Cultural Perspectives

Andrea Ravignani 1,2,3 , Tania Delgado 4 , Simon Kirby 5 , Nori Jacoby 6 , Josh McDermott 7 , Jessica A. Grahn 8 , Daniel J. Cameron 6 , Aniruddh D. Patel 9,10 , Ryuji Takeya 11 , Masashi Kameda 11 and Masaki Tanaka 11

1Veterinary and Research Dept., Seal Centre Pieterburen, The Netherlands

2Max Planck Institute for Psycholinguistics, The Netherlands

3AI-Lab, Vrije Universiteit Brussel, Belgium

4Department of Cognitive Science, University of California, San Diego, USA

5Centre for Language Evolution, School of Philosophy, Psychology and Language Sciences, University of Edinburgh, UK

6The Center for Science and Society, Columbia University, USA

7Department of Brain and Cognitive Sciences, Massachusetts Institute of Technology, USA

8Brain and Mind Institute and Department of Psychology, Western University, Canada

9Department of Psychology, Tufts University, USA

10Azrieli Program in Brain, Mind, and Consciousness, Canadian Institute for Advanced Research (CIFAR), Canada

11Department of Physiology, Hokkaido University School of Medicine, Japan

Symposium organizer: Andrea Ravignani

Musical rhythm is a ubiquitous feature of the human species (Patel, 2010). Around the world, rich rhythmic and musical structures originate from complex interactions between biological and culturally determined mechanisms (Fitch, 2017). Cross-cultural and cross-species research offer complementary roles in the understanding of the origin of musical rhythm: the former can provide evidence for features that differ from culture to culture, whereas the latter point to biologically determined mechanisms that are shared between humans and other animals. In the past few years, paradigms from psychology and music cognition that were previously run predominantly on North American and European subjects have been increasingly incorporated within cross-cultural settings. At the same time, music cognition paradigms have also been increasingly adapted to animal research, providing striking evidence that, given appropriate training, some animals can successfully complete tasks previously considered specific to humans. The crucial bottleneck limiting progress in both fields is the relative lack of paradigms that can be applied in wider experimental contexts (Trehub, 2015). Moreover, three approaches in research on timing have been historically separated. First, the psychology of timing and time perception have been studied in meticulously controlled laboratory experiments. Second, temporal patterns in animal behavior have been investigated by pure behavioral observation, with lack of interest to connect their findings to human cognition. Third, study of the temporal dimension of music, i.e., rhythm, has fallen in the domain of the humanities (classical musicology) and social sciences (cultural anthropology). This symposium showcases work that traverses these domains. Two of the papers describe implementations of a recent experimental technique known as “iterated learning” to rhythm. In iterated learning, participants are asked to reproduce random rhythms; their reproductions are fed back as the stimulus, forming a chain of stimuli that are increasingly dominated by cognitive and productive biases. Ravignani and colleagues use the technique to simulate cultural transmission in the lab, featuring miniature societies of interacting individuals in a controlled environment (and hinting at possible connections with natural animal timing). Jacoby and McDermott use the same technique to extract internal representation of rhythms, showing that the technique can be successfully applied on individuals without musical experience and from a wide range of cultural backgrounds. Grahn and Cameron show how EEG- based music cognition paradigms can be adapted to reveal implicit culturally dependent rhythm processing in individuals from Canada and Africa, while Patel and colleagues show how eye movements can produce predictive and tempo-flexible synchronization in monkeys, behaviors that were recently regarded as human specific. Taken together, these works reveal that challenges and promise of empirically studying the origin of rhythm from both cross-cultural and cross-species perspectives.


Rhythm, music, beat, comparative, cross-cultural, cross-species

1.Links Predictive and Tempo-Flexible Synchronization to a Visual Metronome in Monkeys (A. D. Patel, R. Takeya, M. Kameda and M. Tanaka)

Predictive and tempo-flexible synchronization to an auditory beat is a fundamental component of human music. Prior research training macaque monkeys to tap to an auditory or visual metronome has found their movements to be largely reactive rather than predictive. Does this reflect the lack of a capacity for predictive synchronization in monkeys, or a lack of motivation to exhibit this behaviour? To discriminate these possibilities, we trained monkeys to make synchronized eye movements to a visual metronome. We found that the monkeys could generate predictive saccades synchronized to periodically alternating visual stimuli when an immediate reward was given for every predictive response. This behaviour generalized to novel untrained tempi, and the monkeys could maintain the tempo internally (i.e., even when the target was triggered by saccades). These results suggest that monkeys have the capacity for predictive synchronization to a visual beat, but are not intrinsically motivated to do it.

2.Cross-Cultural Comparisons of Neural and Motor Entrainment to the Beat (J. A. Grahn and D. J. Cameron)

Although all human cultures have rhythm, the rhythmic structures, and perception of those structures, differ across cultures. The current study investigates how culture influences behavioural and neural processing of auditory rhythms. Participants from two cultural groups (San and Hambukushu) from different regions of Botswana, and a group from North America, listened passively to San, Hambukushu, and Western rhythms during encephalographic (EEG) recording. All groups also tapped the beat of the rhythms at a separate time. We predicted that neural and behavioural measures would differ between the cultural groups depending on rhythmic familiarity, with a positive relationship between beat tapping and neural entrainment to particular beat rates. Preliminary analysis suggests cultural differences in both neural and motor entrainment to the beat of musical rhythms, with effects of cultural familiarity present. These novel data provide evidence of how musical rhythm is perceived and entrained to in distinct cultures.

3.Rhythmic Perceptual Priors Revealed Cross-Culturally by Iterated Reproduction (N. Jacoby and J. McDermott)

Probability distributions over external states (priors) are essential to the interpretation of sensory signals. We propose a novel method to characterize perceptual priors and apply it to the study of simple rhythms. Simple rhythms have been previously studied using discrimination, categorization or production paradigms. We argue that results obtained using these methods can be re-interpreted as manifesting subjects’ reliance on perceptual priors. We developed a method based on iterated reproduction of random temporal sequences. Listeners were asked to reproduce random “seed” rhythms; their reproductions were fed back as the stimulus, and eventually became dominated by internal biases, such that priors could be estimated by applying the procedure multiple times. By testing the paradigm amongst US participants with different levels of musical expertise as well as members of the Tsimane, a native Amazonian society with limited exposure to Western music, we show that priors are strongly modulated by culturally-dependent musical exposure.

4.The Evolution of Rhythm between Biology and Culture (A. Ravignani, T. Delgado and S. Kirby)

Musical rhythm, beyond its variety, exhibits cross-cultural similarities and statistical universals. Testing the mechanisms underlying these universals, I will show human experiments where musical rhythm is created and evolves culturally due to cognitive and motoric biases. I will also suggest how comparative animal research can help reconstruct early hominid musicality.


Fitch, W. T. (2017). Cultural evolution: Lab-cultured musical universals. Nature Human Behaviour, 1, 0018.

Patel, A. D. (2010). Music, language, and the brain. Oxford: Oxford University Press.

Trehub, S. E. (2015). Cross-cultural convergence of musical features. PNAS, 112, 8809–8810.

Timing and Conditioning: A Contemporary Overview

Domhnall Jennings 1 , Charlotte Bonardi 2 , David J. Sanderson 3 , Joseph M. Austen 3 , Rolf Sprengel 4 , and Charles Randy Gallistel 5

1CEA1Institute of Neuroscience, Newcastle University, Newcastle upon Tyne, UK

2School of Psychology, University of Nottingham, University Park, Nottingham, UK

3Department of Psychology, Durham University, Durham, UK

4Max-Planck Institute of Medical Research, Department of Molecular Neurobiology, Germany

5Rutgers Centre for Cognitive Science, Psych Bldg Addition, Busch Campus, NJ, USA

Symposium organizers: Domhnall Jennings and Charlotte Bonardi

Timing has always occupied a central place in conditioning theory: the conditioning process is engaged by events that are temporally contiguous, and conditioning speed is profoundly affected by temporal factors. Moreover, conditioning tasks result in timing: animals will learn not only that the conditioned stimulus (CS) predicts the unconditioned stimulus (US), but also when that US will occur. But the close relationship between these two phenomena is not mirrored in the theories proposed to explain them. Theories of conditioning and timing are focussed on explaining different aspects of the learned behaviour, often using different behavioural measures. As a consequence most conditioning theories typically offer little explanation of timing, while established timing theories often struggle to account for the broad range of phenomena revealed in conditioning studies (Kirkpatrick, 2014). More recent work has begun to bridge this theoretical chasm, and both timing and conditioning theories have been adapted to encompass a much broader range of effects. In addition more effort has been devoted to exploring the interaction between conditioning and timing manipulations on behaviour. The objective of this symposium is to provide a brief contemporary overview of this research. Bonardi and Jennings’ work aims to explore the effect of temporal factors on conditioning, and of conditioning manipulations on timing. Jennings and Bonardi report the results of experiments exploring the effect of latent inhibition training – preexposure of the CS prior to CS-US pairings, which typically retards acquisition of conditioned responding – on timing ability. Their findings suggest that CS preexposure both retards conditioning but enhances timing accuracy. Bonardi and Jennings describe the results of a trace conditioning task, in which a fixed-duration CS is followed by the US after a trace interval which could either be of fixed or variable duration. The results suggest that the CS acquired greater associative strength when the trace interval was variable. Sanderson and Austin’s presentation focusses on the neural bases of conditioning and timing behaviour, in a report of work with mice lacking the GluA1 subunit of the AMPA receptor. These animals fail to show the normal attenuation of learning that is seen with longer-duration CSs. In an elegant series of experiments they demonstrate that this is due to a lack of sensitivity to rate of reinforcement, accompanied by normal sensitivity to number of CS/US pairings. Gallistel’s contribution represents the information-processing approach to explaining both conditioning and timing effects. This eschews conditioning theory’s usual focus on the trial (a CS-US pairing) in favour of a more global and mathematical analysis of the prevailing environmental conditions. In this presentation Gallistel focusses on the phenomenon of inhibitory learning, in which a CS signals the omission of an otherwise expected US (No-US). The trial-based approach of associative theories struggles to explain events that have not occurred, and Gallistel explores the possibility that this uncertainty might be more fruitfully examined by applying a hazard function to the probability that a reinforcing event will occur.


Conditioning, timing, information, animals

1.Temporal Control of the Conditioned Response Following CS-Preexposure: Retarded Learning Does Not Mean Retarded Timing (D. Jennings and C. Bonardi)

When a conditioned stimulus (CS) is paired with an unconditioned stimulus (US), subjects learn not only that the US will be delivered (conditioning), but also when (timing). This learning can be attenuated in some paradigms, such as latent inhibition (LI) in which preexposure of the CS retards learning when CS and US are subsequently paired. However, little is known about the effect of CS preexposure on timing. In a series of experiments using rats in an appetitive procedure, we investigated whether latent inhibition was evident in timing behaviour. Results showed that LI was evident as a lower rate of conditioned responding to the preexposed CS than to a novel cue. Paradoxically, response slopes calculated across the duration of the CS were higher for the preexposed CS: thus, temporal control was better following preexposure. These results indicate that although CS preexposure retards conditioning, it may enhance timing.

2.Rate-Sensitive Learning Requires the GluA1 Subunit of the AMPA Receptor (D. J. Sanderson, J. M. Austen and R. Sprengel)

Conditioning occurs more readily with cues of short duration than of long duration. This effect is abolished by deletion of the GluA1 subunit of the AMPA glutamate receptor in mice. The cue duration effect can also be abolished by reinforcing the short duration cue at the same rate as the long duration cue (e.g., a 10 s cue reinforced on 25% of trials and a 40 s cue reinforced on 100% of trials), suggesting that sensitivity to reinforcement rate is the cause of the cue duration effect

in control mice. GluA1 knockout mice are insensitive to reinforcement rate and instead are sensitive to the number of times that a cue is paired with reinforcement, independent of whether rate is manipulated by cue duration or probability of reinforcement per trial. These results suggest that GluA1 is necessary for weighting numeric information by temporal information in order for rate-sensitivity to be achieved.

3.The Bernoulli-Gauss and the No-US (C. R. Gallistel)

A fundamental problem in the theory of associative learning is the specification of the time at which No-USs “occur”. No-USs are events that fail to occur. How can an event that doesn’t occur have a time of occurrence? Wilkes and Gallistel propose that the brain represents event times with one of two maximum entropy stochastic models: the exponential and the Bernoulli-Gauss. The Bernoulli-Gauss marries the Bernoulli distribution to the Gaussian distribution to produce a Bernoulli distribution situated in time. Its hazard function provides a natural basis for localizing in time both expected events and expected failures. The hazard for both occurrence and failure to occur rise and subside, demarcating regions in time at which both are anticipated with complementary probabilities.

4.The Effects of Stimulus Distribution Form on Trace Conditioning (C. Bonardi and D. Jennings)

Two groups of rats were conditioned to a fixed duration CS in a trace interval task, in which a trace interval separates CS offset and US delivery. For rats in Group Fix the trace interval was of a fixed duration, while for rats in Group Var its duration varied. In contrast to our findings in delay conditioning tasks (where US delivery occurs at CS offset) responding during the CS was higher in Group Var than in Group Fix. However, during the trace interval this difference was reversed – Group Fix responded at a higher rate than Group Var. A further experiment employed a blocking procedure to examine whether the greater response rate observed during the CS in Group Var was due to a performance effect, or the acquisition of greater associative strength by the CS. The results supported the second interpretation. The theoretical implications of these findings will be discussed.


Kirkpatrick, K. (2014). Interactions of timing and prediction error learning. Behav. Processes, 101, 135–145.

Time Processing Deficits in Developmental Disorders

Laurel J. Trainor 1,2,3 , Andrew Chang 1 , Jennifer Chan 1 , Yao-Chuen Li 4,5 , John Cairney 4,6 , B. Tillmann 7 , L.-H. Canette 7,8,9 , N. Bedoin 9 , Usha Goswami 10 , Valdas Noreika 11 and Christine Falter 12

1Department of Psychology, Neuroscience and Behaviour, McMaster University, Hamilton, ON, Canada

2McMaster Institute for Music and the Mind, McMaster University, Hamilton, ON, Canada

3Rotman Research Institute, Baycrest Hospital, Toronto, ON, Canada

4Department of Kinesiology and Infant and Child Health (INCH) lab, Department of Family Medicine, McMaster University, Canada

5Child Health Research Center, Institute of Population Health Sciences, National Health Research Institutes, Taiwan

6Faculty of Kinesiology & Physical Education, University of Toronto, Canada

7Lyon Neuroscience Research Center CNRS-UMR 5292, INSERM U1028, University Lyon, France

8LEAD-CNRS 5022, Université de Bourgogne, France

9Dynamique Du Langage Laboratory CRNS-UMR 5596, University Lyon, France

10Centre for Neuroscience in Education, University of Cambridge, UK

11Department of Psychology, University of Cambridge, UK

12Institute of Medical Psychology, Ludwig-Maximilians-University, Germany

Symposium organizer: Laurel J. Trainor

Time processing is essential for perception, cognition and action, as well as coordinating between activities and different brain processes. Although most research on the major development disorders, including Dyslexia, Autism, Attention Deficit and Hyperactivity Disorder (ADHD), and Developmental Coordination Disorder (DCD), has focused on unique features of each disorder, there is high comorbidity among them, suggesting that there may be common underlying deficits. One candidate involves deficits in time processing. Studies to date have shown that children with these developmental disorders are more likely to have inferior perceptual timing (e.g., discriminating temporal intervals or rhythms, judging temporal order of events) and/or motor timing (e.g., coordinating actions in time, reproducing auditory rhythms; see Falter & Noreika, 2014, for a review). However, the commonality of timing deficits across developmental disorders remain far from clear due to the lack of common test paradigms, documented developmental trajectories, and an integrative theory. It appears that Autism involves impaired social timing, Dyslexia involves impaired phonological processing and temporal sequencing of syntax, DCD involves impaired sensorimotor timing, and ADHD involves impaired temporal executive and attentional control. However, the extent to which basic time processing is commonly impaired is not known. The objectives of this symposium are to assess this idea through talks on timing deficits in Dyslexia, Autism, ADHD and DCD that include behavioural and neural measures. The efficacy of timing interventions (e.g., auditory rhythmic cue) in Parkinson’s disease and Dyslexia will be discussed, as well as whether such interventions could be successfully applied to other disorders.


Patient, 100s of ms-secs, behaviour, electrophysiology, perceptual and motor timing

1.Auditory Timing Deficits in Developmental Coordination Disorder (A. Chang, J. Chan, Y.-C. Li, J. Cairney and L. J. Trainor)

Developmental Coordination Disorder (DCD) is a neurodevelopmental disorder with deficits in motor coordination. Prior research suggests that perceiving auditory timing may involve cortical networks that include motor areas. However, it remains unclear whether the motor system is necessary for perceiving time. If this is the case, children with probable DCD should have inferior temporal discrimination performance compared to typically developing (TD) children. In Experiment 1, we investigated discrimination thresholds for auditory unfilled duration and rhythm. The results showed that children with probable DCD have larger thresholds for unfilled duration than TD children, but are not significantly different on rhythm. In Experiment 2, electroencephalogram was recorded while participants listened to auditory oddball sequences with deviations in unfilled duration or rhythm. Analyses on mismatch negativity, which reflects preattentive perceptual encoding, are ongoing. Together, the results suggest auditory and motor systems are connected for perceiving time, and suggest potential auditory interventions for DCD.

2.Temporal Processing Deficits and Benefits of Rhythmic Auditory Stimulation on Syntax Processing in Developmental Language Disorders (B. Tillmann, L.-H. Canette and N. Bedoin)

Children with developmental language disorders have been shown to be impaired not only in language processing, but also in rhythm and meter perception. We tested the influence of external rhythmic auditory stimulation (i.e., musical rhythms) on syntax processing in children with specific language impairment and children and adults with dyslexia, using behavioral and electrophysiological measurements. Grammaticality judgments for auditorily presented (correct or incorrect) sentences were better after regular musical prime sequences than after irregular sequences or baseline sequences. In addition, the P600, an electrophysiological marker for processing grammatical errors, was enhanced after regular prime sequences. We also collected data for temporal processing in dyslexic adults. Our findings are interpreted within the Dynamic Attending Theory (Jones, 1976) and the Temporal Sampling (oscillatory) Framework for developmental language disorders (Goswami, 2011). They encourage the use of rhythmic structures (even in non-verbal materials) to boost linguistic structure processing and outline perspectives for rehabilitation.

3.Reading, Rhythmic Timing and the Brain (U. Goswami)

Recent neural studies of speech processing provide a temporal “oscillatory” perspective on brain mechanisms for speech encoding. Using these insights, I develop an oscillatory “temporal sampling” neural framework for linking the auditory processing of rhythm to phonological development in dyslexia (Goswami, 2011). I show that for English, sensitivity to temporal rhythmic structure is core to developing good phonological skills, and that English children with dyslexia are relatively insensitive to rhythmic timing. Rhythmic sensitivity is related to how efficiently the brain processes the amplitude modulation (energy) patterns in speech. The rhythmic energy patterns in speech occur at multiple temporal rates simultaneously (delta, theta, beta, gamma). I will describe neural studies showing atypical rhythmic entrainment for dyslexic readers in English.

4.Time Processing in ASD and ADHD: Shared Deficit or Disorder-Specific Abnormalities? (V. Noreika and C. Falter)

Individuals with attention-deficit/hyperactivity disorder (ADHD) and autism spectrum disorder (ASD) have been reported to have a range of abnormalities in motor timing, perceptual timing, and time perspective functions. However, there is a lack of studies directly comparing time processing between individuals with ADHD, ASD, and typically developing controls. Arguably, three hypotheses should be formally considered when contrasting any two disorders known to have timing abnormalities. A “unification” hypothesis would suggest that ADHD and ASD have a shared timing deficit with similar if not identical timing abnormalities. A “differentiation” hypothesis would propose that ADHD and ASD have disorder-specific timing abnormalities. A “mixed” hypothesis would imply that some of the timing abnormalities are shared between ADHD and ASD, while other timing deficits are disorder-specific. Systematic literature review as well as our own original studies seem to provide preliminary support for the third “mixed” hypothesis.


Allman, M., & Falter, C. (2015). Abnormal timing and time perception in autism spectrum disorder? A review of the evidence. In A. Vatakis & M. Allman (Eds.) Time Distortions in Mind–Temporal Processing in Clinical Populations. Leiden: Brill.

Falter, C. M., & Noreika, V. (2014). Time processing in developmental disorders: A comparative view. In V. Arstila & D. Loyd (Eds.) Subjective Time: The Philosophy, Psychology, and Neuroscience of Temporality. Cambridge, MA: MIT Press.

Goswami, U. (2011). A temporal sampling framework for developmental dyslexia. Trends Cogn. Sci., 15, 3–10.

Temporal Organization of Perceptual Processes by Motor-Driven Low-Frequency Neuronal Oscillations

Benjamin Morillon 1 , Alice Tomassini 2 , Pieter Medendorp 2 , Eric Maris 2 and Daniele Schon 1

1Aix Marseille Univ, Inserm, INS, Institut de Neurosciences des Systèmes, Marseille, France

2Donders Institute for Brain, Cognition and Behavior, The Netherlands

3Auditory Language Group, Department of Basic Neuroscience, University of Geneva, Switzerland

Symposium organizer: Benjamin Morillon

A consistent body of research in the last decade has highlighted the role of oscillatory activity in sensory processing, leading to the notion that perception is inherently discrete and periodic. More recently, the focus has also turned to the role of motor processes in the proposed periodic nature of perception. This goes far beyond the long- standing idea that externally-triggered movements might be dominated by rhythmic components. The motor system might actually be capable of exerting endogenous control of oscillatory activity that entails perceptual consequences. Motor-related modulations of perceptually-relevant oscillations have been shown when sensory information is predictable in time – suggesting that the motor system orchestrates the temporal tuning of attention, optimizing information extraction during active exploration (Morillon et al., 2014; Arnal et al., 2015). Furthermore, oscillations in perceptual performance time-locked to the execution of voluntary movements have been shown for stimuli that are unpredictable and, importantly, unrelated to the motor task, pointing to a rather automatic form of sensory-motor coupling (Tomassini et al., 2015). In this symposium, we will present novel findings that reveal a leading role of the motor system in the temporal prediction of external events and the modulation of sensory processing through rhythmic brain activity, and discuss their implications for action-perception coupling mechanisms.


Prediction, rhythms, behavior, MEG, adult, 100s of ms-secs

1.Theta Oscillations Synchronize Perception with Motor Intention (A. Tomassini, P. Medendorp and E. Maris)

Our motor system orchestrates the sampling of sensory information by orienting our receptor organs in space and time. Mounting evidence further suggests that motor signals also contribute to the actual analysis of the incoming sensory data, thereby shaping the perceptual outcome. I will present behavioral and neurophysiological findings showing that the motor system can modulate low-level sensory function, even when the movement does not involve the receptor organ and the stimuli are irrelevant for the motor performance. Crucially, this motor-related modulation has a rhythmic signature, occurs in an anticipatory fashion – i.e., during movement planning - and entails perceptual consequences. We found rhythmic fluctuations of visual perception, which are time-locked to the execution of voluntary hand movements and emerge before movement onset. This movement-locked rhythmicity in perceptual performance is predicted by the phase of EEG theta oscillations (~4 Hz) very early during movement planning (>1 s before movement). Moreover, theta oscillations are phase- locked to the onset of the movement. Remarkably, the alignment of theta phase and its perceptual relevance unfold with similar non-monotonic profiles, indicating their relatedness. Action planning hence seems to be accompanied by an endogenous phase adjustment of perceptually-relevant brain oscillations. The present work suggests that action and perception may be bound in an automatic way since the very early processing stages through neuronal oscillatory activity in the theta range.

2.Motor Origin of Temporal Predictions in Auditory Attention (B. Morillon)

Temporal predictions are fundamental instruments for facilitating sensory selection, allowing humans to exploit regularities in the world. It is proposed that the motor system instantiates predictive timing mechanisms, helping to synchronize temporal fluctuations of attention with the timing of events in a task-relevant stream. I will present a neurophysiological account for this theory in a paradigm where participants track a slow reference beat while extracting auditory target tones delivered on-beat and interleaved with distractors. At the behavioral level I will show that overt rhythmic movements sharpen the temporal selection of auditory stimuli, thereby improving performance. Capitalizing on magnetoencephalography recordings I will provide evidence that temporal predictions are reflected in Beta-band (~20 Hz) energy fluctuations in sensorimotor cortex and modulate the encoding of auditory information in bilateral auditory and fronto-parietal regions. Together, these findings are compatible with Active Sensing theories, which emphasize the prominent role of motor activity in sensory processing.

3.Music to Speech Entrainment (D. Schon)

The Dynamic Attending theory suggests that attention is entrained to the rhythmic structure of music via stimulus brain oscillatory coupling. Thus, different point in time will be differently attended. Indeed, stimuli presented at attended points in time, corresponding to the strong metrical beats of music, are processed faster and more accurately compared to stimuli presented out of beat or on weak beats. This holds true when the target stimulus is presented in the visual modality. Importantly this effect last after the rhythmic stimulus end, showing a carry-over effect of entrainment in time. Recent work demonstrates that a rhythmic stimulus can facilitate the processing a of a following speech stimulus. I will provide evidence of the mechanism underlying this music to speech coupling as well as the functional relevance in terms of speech processing and speech rehabilitation.


Arnal, L. H., Doelling, K. B., & Poeppel, D. (2015). Delta-beta coupled oscillations underlie temporal prediction accuracy. Cerebral Cortex, 25, 3077–3085.

Morillon, B., Schroeder, C. E., & Wyart, V. (2014). Motor contributions to the temporal precision of auditory attention. Nat. Comm., 5, 5255.

Tomassini, A., Spinelli, D., Jacono, M., Sandini, G., & Morrone, M. C. (2015). Rhythmic oscillations of visual contrast sensitivity synchronized with action. J. Neurosci., 35, 7019–7029.

Neural Oscillations for Time Estimation

Martin Wiener 1 , Nandakumar Narayanan 2 , Tadeusz W. Kononowicz 3 , Clemence Roger 3 , Virginie van Wassenhove 3 , Alomi Parikh 4 , Arielle Krakow 5 and H. Branch Coslett 4

1George Mason University, USA

2University of Iowa, USA

3CEA/DRF/NeuroSpin, INSERM Cognitive Neuroimaging Unit, France

4University of Pennsylvania, USA

5Johns Hopkins University, USA

Symposium organizer: Martin Wiener

The perception of time and prediction of upcoming events requires coordination between a diverse set of neural regions. Further, interval timing may be fractionated into separate yet overlapping neural circuits that are invoked for different temporal contexts across the brain (i.e., rhythmic vs. non-rhythmic). Neural oscillations have emerged as a candidate mechanism for neural timing and the coordination of activity across different timing contexts and brain regions (Wiener & Kanai, 2016). Across the assemblage of frequency bands, numerous associations have been made between distinct frequency bands and timing functions in particular task contexts. For example, delta oscillations (1–4 Hz) have been observed coordinating frontal and cerebellar circuits during interval timing (Parker et al., 2017), whereas beta oscillations (15–25 Hz) have recently been associated with supra-second temporal reproduction (Kononowicz & van Rijn, 2015). In this symposium, we will provide an overview of various candidate oscillations and speculate on their precise function for interval timing. The work presented in this symposium will span both human and animal models, and include a variety of methods to provide convergent findings.


Oscillations, perceptual timing, noninvasive brain stimulation, decision making, accumulation

1. Delta Oscillations and the Starting Gun (N. Narayanan)

Timing processes can be initiated by cues. However, how these cues engage neuronal timing mechanisms is unclear. Here, we present evidence that during fixed-interval timing, cues trigger delta- frequency (1–4 Hz) oscillations in medial frontal cortex of humans and rodents, These rhythms require medial frontal D1 dopamine receptors. We show that this rhythm can engage single neurons in the rodent striatum and subthalamic nucleus and the human subthalamic nucleus that are involved in timing. Furthermore, stimulating medial frontal D1 dopamine-receptor expressing neurons at delta frequencies can compensate for depleted frontal dopamine. These data indicate that D1-dependent delta oscillations may play a role in initiating the neural mechanisms controlling timing.

2. Temporal Metacognition as Decoding Self-Generated Brain Dynamics (T. W. Kononowicz, C. Roger and V. van Wassenhove)

In The awareness of one’s sources of errors and misjudgments is key to cognitive and behavioral improvement. People can self-assess various types of cognitive performance when decisions involve external perception. However, without external feedback, how can internal timing and errors be at once generated by the same brain? Strikingly, we show that humans’ ability to generate a duration and to self-assess their timing rely on an internal variable set by the synchronization of beta oscillations. Using time-resolved neuroimaging, we show that the strength of beta synchronization informs on how distant from a set internal variable one’s time production is. Beta synchronization sets a ballistic trajectory for time production, which is read-out and integrated in the performance monitoring subsystems as alpha desynchronization following the production of duration, enabling awareness of timing errors. Altogether, our study provides novel insights on how the human brain may decode purely internal dynamics during temporal metacognition.

3. Causal Role of Beta Oscillations in Time Estimation (M. Wiener, A. Parikh, A. Krakow and H. B. Coslett)

Recent evidence has suggested that beta oscillations (15–25 Hz) serve as a supramodal index of timing functions in the brain. Yet, no study has investigated whether beta oscillations are causally related to timing functions. To provide a comprehensive overview of beta’s involvement in timing, we first re-analyzed two prior EEG datasets (Wiener et al., 2012; Wiener & Thompson, 2015) from perceptual timing paradigms, and observed changes in beta associated with the encoding of durations into memory. Next, we tested 20 human subjects with transcranial alternating current stimulation (tACS) over the supplementary motor area at alpha (10 Hz) and beta (20 Hz) frequencies during a temporal bisection task, and found that beta stimulation exclusively shifts the bisection point, while preserving precision. Finally, we decomposed behavioral data with a drift diffusion model, finding that the shift in perception can be tied to a change in the starting point of evidence accumulation.


Kononowicz, T. W., & van Rijn, H. (2015). Single trial beta oscillations index time estimation. Neuropsychologia, 75, 381–389.

Parker, K. L., Kim, Y. C., Kelley, R. M., Nessler, A. J., Chen, K.-H., Muller-Ewald, V. A., Andreasen N. C., & Narayanan, N. S. (2017). Delta-frequency stimulation of cerebellar projections can compensate for schizophrenia-related medial frontal dysfunction. Mol. Psychiatry., 22, 647–655.

Wiener, M., & Kanai, R. (2016). Frequency tuning for temporal perception and prediction. Current Opinion in Behavioral Sciences, 8, 1–6.

Timing, Neural Dynamics, and Temporal Scaling

Dean Buonomano 1 , Hugo Merchant 2 and Mehrdad Jazayeri 3

1Departments of Neurobiology and Psychology, UCLA, USA

2Laboratorio B-15, Instituto de Neurobiología - UNAM, Querétaro, México

3McGovern Institute, Robert A. Swanson Career Development Professor in the Life Sciences, Brain and Cognitive Sciences MIT, USA

Symposium organizer: Dean Buonomano

There is increasing experimental and theoretical support for the notion that timing on the scale of hundreds of milliseconds to second relies on neural dynamics—i.e., changing populations of neural activity, in which time is encoded in the population of neurons active at any given movement (population clocks). Electrophysiological recordings in a large number of brain areas, including the motor cortex, premotor cortex, prefrontal cortex, hippocampus, and striatum have reported that it is possible to decode elapsed time from reproducible patterns of neural activity in these areas. A number of questions, however, remain unclear, including the neural mechanisms underlying temporal scaling, that is, the ability to produce the same motor patterns at different speeds. This symposium will explore behavioral and neurophysiological data, and neurocomputational models, underlying timing and temporal scaling in the range of hundreds of milliseconds to seconds.


Timing, neural dynamics, neurophysiology, neurocomputation

1. Periodic Neural State Trajectories Underlie Rhythmic Tapping (H. Merchant)

The ability to generate rhythms of different tempos is a hallmark of higher cognition. We know that tapping to a regular beat engages neurons from the medial premotor cortices (MPC). Yet, the neuronal population code behind rhythmic tapping remains elusive. Here we found that the activity of hundreds of primate MPC neurons show a strong periodic pattern that becomes evident when its activity is projected into a lower dimensional state space. We show that different tempos are encoded by circular trajectories of different radii and that this neuronal code is highly resilient to the number of participating neurons. Crucially, the oscillatory dynamics in neuronal state space is a signature of cognitive timing under metronome guidance or when is internally controlled, and is not the result of repetitive motor commands. Our results support the notion that rhythmic behaviors are encoded by the dynamic state of MPC neural populations.

2. Flexible Temporal Control of Self-Initiated Movements by Speed of Cortical Dynamics (M. Jazayeri)

The medial frontal cortex plays a fundamental role in temporal control of movements and has been implicated in inhibition, initiation, and coordination of voluntary actions. However, the underlying computational principles and neural mechanisms remain largely unknown. We recorded from the medial frontal cortex of monkeys trained to flexibly produce different intervals with different effectors. Neurons displayed complex and heterogeneous activity patterns. Despite this complexity, response profiles of many neurons were similar when scaled to a reference interval. This property indicates that the produced interval can be understood in terms of the speed of response modulations across the population along invariant neural trajectories. A recurrent neural network emulating this behavior revealed that (1) speed is regulated by an external input exploiting single-neuron nonlinearities, and (2) invariant trajectories are governed by recurrent interactions controlling dynamics. Together, these findings reveal the coding principles and potential mechanisms by which cortical dynamics flexibly control self- timed voluntary movements.

3. Temporal Scaling of Complex Temporal Patterns in a Recurrent Neural Network Model and Humans (D. Buonomano)

Timing is fundamental to complex motor behaviors: from tying a knot to playing the piano. A general feature of motor timing is temporal scaling: the ability to produce similar motor patterns at different speeds. Here we report that temporal scaling is not an intrinsic property of motor timing: after learning to tap a Morse code signal subjects were not able to accurately generate it at faster and slower speeds. We then demonstrate that while temporal scaling is also not a general property of recurrent neural networks (RNNs), that RNNs can be trained to produce robust temporal scaling. The model captures a signature of motor timing—Weber’s law—but predicts that timing will be more precise at faster speeds. This prediction was confirmed in a temporal production task: the standard deviation of a response at the same absolute time varied across speeds.

Perception: Continuous or Discrete?

Michael Herzog 1 , Adrien Doerig 1 , Rufin VanRullen 2 , Tomer Fekete 3 and Cees van Leeuwen 3

1EPFL, Switzerland

2CerCo, CNRS, Universite de Toulouse, France

3KU Leuven, Belgium

Symposium organizers: Michael Herzog and Adrien Doerig

Intuitively, consciousness seems to be a continuous stream of percepts. A diver is jumping off a cliff. We see her trajectory against the blue sky at each single moment in time. Many models in vision rely explicitly and implicitly on continuous perception. For example, in visual backward masking, continuous perception is assumed to explain how a trailing mask can render a preceding target unconscious. However, continuous accounts are in fact challenged by demonstrations where, as in masking, a trailing element determines perception. Another classic example is the color phi paradigm. When a green disk is presented for a short time, it is perceived as such. When a second red disk is presented a bit later and at a different location, motion is perceived during which the color changes midway. Perception has changed qualitatively. Whereas discrete models have been proposed for centuries, continuous perception seems to be so intuitive that there are only a few explicit proponents. In this symposium Cees van Leeuwen and Tomer Fekete (Leuven) will propose and defend the continuous account against discrete proposals. Rufin VanRullen (Toulouse) will show how brain rhythms produce discrete perception, which can be envisioned as a one-stage model of discrete perception. Michael Herzog and Adrien Doerig (Lausanne) will propose a two-stage model of discrete perception, in which the temporal resolution of unconscious processing is independent of the duration of percepts, which can last up to 400 ms. Clearly, the temporal nature of perception is one of the fundamental questions in perception research and, as mentioned, the answer to this question has important ramifications for most fields of perception science. For this reason, it is surprising to see this topic remain largely unexplored. This controversial symposium is the first step in understanding the temporal nature of perception in a systematic fashion.


Consciousness, resolution, modeling, adult, 10s-100s of ms

1. Perceptual Cycles (R. VanRullen)

Brain function involves oscillations at various frequencies. This could imply that perception and cognition operate periodically, as a succession of cycles mirroring the underlying oscillations. This age-old notion of discrete perception has resurfaced in recent years, fueled by advances in neuroscientific techniques. Contrary to earlier views of discrete perception as a unitary sampling rhythm, contemporary evidence points not to one but several rhythms of perception that may depend on sensory modality, task, stimulus properties, or brain regions. In vision for example, a sensory alpha rhythm (~10 Hz) may co-exist with at least one more rhythm performing attentional sampling around 7 Hz. Although these timescales are clearly slower than the temporal resolution observed for many visual features (e.g., 30–40Hz for motion or flicker), there is no contradiction as long as discrete sampling takes place after these features have been extracted from the sensory input stream.

2. Two Steps: Quasi-Continuous Unconscious Processing and Discrete Conscious Perception (M. Herzog and A. Doerig)

Perception seems to be a continuous stream and, for this reason, we implicitly assume that perception is continuous. However, we need to integrate information across time. For example, in apparent motion we do not perceive first a static disk and then another static disk but a smooth trajectory, favoring discrete accounts of perception. What is the sampling rate of discrete perception? Usually, the sampling rate is determined by temporal resolution. If we cannot perceive two flashes of light presented 40 ms after each other, discrete sampling cannot be faster than 40 ms. However, different paradigms have shown evidence for sampling rates ranging from 3 to 300 ms. Here, we propose that the sampling rate in these paradigms is determined by the temporal resolution of unconscious processing, which is independent of the changes of conscious percepts, occurring at a much lower rate, likely in the range of 400 ms, as supported by TMS experiments.

3. For the Sake of Maintaining Continuity (T. Fekete and C. van Leeuwen)

For perception to be discrete, neuronal networks must switch states fast enough for us to see, e.g., objects moving. Our data suggest that even for local brain networks this already would render discrete switching practically indistinguishable from continuous progression. In fact discrete switching faces even higher speed demands when inter area communication is considered, given neuronal signaling delays, resulting in asynchrony and enforced waiting times. Brains cannot support discrete switching at these rates. Moreover, postulating it turns out to offer no explanatory advantage, not even for the data claimed to support it. Our experience largely appears continuous, and where rapid transitions occur, as in perceptual switching or sensorimotor coordination, the underlying processes are complex and involve long-term dependencies that accumulate gradually. As a theoretical construct, discreetness therefore offers no added value over continuity. In contrast, continuous multiscale dynamics suffers no such drawbacks and naturally accommodates experience as we know it.

Temporal Binding of Actions to Their Effects: Underlying Mechanisms and Implications for Cognition, Perception, and Development

Marc J. Buehner 1 , Teresa McCormack 2,3 , Sara Lorimer 2,3 , Emma Blakey 2,3 and Christoph Hoerl 4

1Cardiff University, School of Psychology, Park Place, Cardiff, UK

2University of Sheffield, Western Bank, Sheffield, UK

3Queens University Belfast, University Road, Belfast, Northern Ireland, UK

4University of Warwick, Department of Philosophy, Coventry, UK

Symposium organizers: Marc Buehner and Teresa McCormack

Temporal binding (TB) refers to the mutual attraction in subjective time between a cause and its effect: Relative to single-event baseline judgments, people’s perception of causal actions and their outcomes systematically shifts in subjective awareness. Specifically, causal actions are perceived relatively later, while their outcomes are perceived relatively earlier – action and outcome attract each other in subjective experience (Haggard et al., 2002). However, TB has also been reported when time perception is probed directly: Verbal estimates and reproductions of intervals between causal actions and their outcomes are reliably shorter than those of control intervals (Humphreys & Buehner, 2010), and psychophysical measures of causal vs non-causal intervals reveal shorter PSEs in the former than the latter case (Nolden et al., 2012). Because the majority of demonstrations of TB deployed intentional action as the critical cause, TB is now increasingly deployed as a convenient proxy measure for sense of agency. Moreover, reduced TB for negative outcomes has been interpreted to reflect reduced sense of agency for actions that bring about negative consequences: For example, reduced TB for penalties delivered to a peer when following an experimenter’s instruction (as opposed to under free will) is taken to show that coercion reduces one’s sense of agency for self-action (Caspar et al., 2016). The robustness and widespread use of TB notwithstanding, the structural underpinnings and wider implications of TB are still only poorly understood. Very little is known about whether it originates from changes to interval timing, event perception, or both, whether it follows from cue-combination or constraint-satisfaction processes, and how it relates to other cognitive processes. This symposium brings together perspectives from experimental and developmental psychology and philosophy to shine light on the mechanisms underlying TB as well as its implications in a wider context, and its usefulness as a proxy for less tangible constructs such as sense of agency, and causal belief.


Duration, behaviour, adults, children, 100s of ms-secs, attentional modulation, temporal binding

1. Temporal Binding and Internal Clocks (M. J. Buehner)

Most accounts of TB interpret it as stemming from delayed awareness of causal actions combined with earlier awareness of their effects – an approach that can simultaneously explain the two different manifestations of TB reported in the literature: shifts in action perception and changes to interval timing. Changes to time perception, however, are traditionally explained in light of internal clock models (e.g., Gibbon et al., 1984), either as modulations of pacemaker speed (clock rate), or timing latency. I will present evidence from a series of studies showing that TB is associated with reduced clock speed during causal intervals. Moreover, this modulation is specific to the cause-effect interval and does not affect the timing of other, concurrent events. These results are difficult to reconcile with cue combination or belief-updating approaches to TB and question whether TB arising from prospective timing preparations is subserved by the same principles as TB arising from event perception or action preparation paradigms.

2. When Causality Shapes the Experience of Time (E. Blakey and S. Lorimer)

Two studies examined the developmental origins of TB. Participants of all ages (6- to 10-year olds and adults) reported that a causal interval between a key-press and an on-screen rocket launch felt shorter than that same interval between a predictive signal and launch. In a stimulus anticipation task, 4- to 11-year olds and adults predicted when an on-screen event would occur. Participants of all ages predicted that the event would occur earlier on trials where they were aware of a causal mechanism leading to the event compared to trials were the event was merely signaled by a predictor. Notably, TB decreased with age. These results demonstrate that children’s temporal experience, like that of adults, is affected by causal representations. The results point to a bidirectional relation between time and causality that exists early in development and persists into adulthood, and may be privileged very early in development.

3. Temporal Binding and the Perception/Cognition Boundary (C. Hoerl)

Temporal experience has become a particular vibrant area of debate in recent philosophy of mind. TB presents a challenge to one model of temporal experience put forward in this context – the view that we can perceive temporal extension and structure simply because our experiences, too, unfold through time and share the same temporal extension and structure. In response, a defender of this view might argue that TB reflects an error in judgment rather than perception (see, e.g., Phillips, 2014, for related arguments). This raises the general issue as to where to draw the boundary between perception and cognition. I will argue that research on TB – specifically on the mechanisms underpinning it – can in fact also help to show how questions about the perception/cognition boundary should be adjudicated in the context of temporal experience.


Caspar, E. A., Christensen, J. F., Cleeremans, A., & Haggard, P. (2016). Coercion changes the sense of agency in the human brain. Curr. Biol., 26, 1–16.

Gibbon, J., Church, R. M., & Meck, W. H. (1984). Scalar timing in memory. Ann. N. Y. Acad. Sci., 423, 52–77.

Haggard, P., Clark, S., & Kalogeras, J. (2002). Voluntary action and conscious awareness. Nat. Neurosci., 5, 382–385.

Humphreys, G. R., & Buehner, M. J. (2010). Temporal binding of action and effect in interval reproduction. Exp. Brain Res., 203, 465–470.

Nolden, S., Haering, C., & Kiesel, A. (2012). Assessing intentional binding with the method of constant stimuli. Conscious. Cogn., 21, 1176–1185.

Phillips, I. (2014). Experience of and in time. Philos. Compass, 9, 131–144.

Temporal Prediction: Dynamics in Single Neurons and Networks

Matthew S. Matell 1 , Krystal L. Parker 2 , Joseph J. Paton 3 , Valérie Doyère 4 , and Dean V. Buonomano 5

1Department of Psychology, Villanova University, USA

2Department of Psychiatry, University of Iowa Carver College of Medicine, USA

3Champalimaud Centre for the Unknown, Portugal

4Paris-Saclay Institute of Neuroscience (NeuroPSI), CNRS-Université Paris-Sud, Orsay, France

5Departments of Neurobiology and Psychology, University of California, Los Angeles, USA

Symposium organizer: Matthew S. Matell

While we have had a good understanding of the primary behavioral characteristics of temporal control for several decades (Gibbon, 1977), the psychological and neural mechanisms that give rise to this critical facet of behavior remain unclear. Indeed, there are a number of different classes of models, including accumulation and decay models, state-dependent network models using oscillators or intrinsic neural dynamics, and sequential-behavior models that all provide excellent accounts of the behavioral data. Similarly, the neural structures that might underlie these processes span the range of dedicated structures such as the striatum and cerebellum, to the intrinsic dynamics of small and large networks that can be exhibited throughout the brain (Ivry & Schlerf, 2008). Deciphering the mechanisms that underlie timing will therefore require investigating and perturbing the activity patterns of neurons in behaving animals (Merchant et al., 2013). This symposium will provide a forum in which four internationally renowned experts will discuss their work investigating neural activity within different brain areas and how their data support a particular framework for understanding temporal perception and control. Krystal Parker (University of Iowa) will provide evidence that frontal and cerebellar networks interact with one another to facilitate temporal control, and that dysfunction in these circuits contributes to schizophrenia-related temporal disorganization. Joseph Paton (Champalimaud Centre for the Unknown) will present activity patterns of the striatum and its midbrain dopaminergic input neurons and will discuss the direct involvement of these neurons in temporal prediction and control. Valérie Doyère (Paris-Saclay Institute of Neuroscience) will discuss time-related alterations in coupled neural activity patterns in the amygdala and striatum as subjects learn to predict the time of an aversive event. Finally, Dean Buonomano (University of California Los Angeles) will argue that the criticality of temporal prediction for adaptive behavior requires mechanisms that can exist throughout the nervous system. Together, this symposium will address whether network size and specialization is critical for temporal prediction, and whether these factors vary as a function of the duration and behavior in question.

1. An Essential Role for the Cerebellum in Suprasecond Timing (K. L. Parker)

Timing is a fundamental cognitive process, which relies on the frontal cortex. Timing is impaired in schizophrenia, which involves abnormalities in the medial frontal cortex. Here we investigate a novel strategy to normalize medial frontal brain activity by stimulating cerebellar projections. We used an interval timing task requiring frontal and cerebellar networks that are disrupted in schizophrenic patients. We report three novel findings. First, patients with schizophrenia had dysfunctional delta rhythms between 1–4 Hz in the medial frontal cortex. Second, we found in animal models that both frontal and cerebellar neurons were modulated during interval timing and had delta-frequency interactions. Finally, delta-frequency optogenetic stimulation of thalamic terminals on lateral cerebellar projection neurons rescued timing performance as well as medial frontal activity in a rodent model of schizophrenia-related frontal dysfunction. These data provide insight into how the cerebellum influences medial frontal networks and the role of the cerebellum in cognitive processing.

2. Basal Banglia Contributions to a Time-Based Decision (J. J. Paton)

Time is a fundamental dimension of experience, critical for extracting meaning from the environment and constructing behavior. However, the neural mechanisms for timing are poorly understood. We trained rodents to judge time intervals as longer or shorter than 1.5 seconds while recording and manipulating activity of neurons the dorsal striatum and dopamine neurons in the substantia nigra pars compacta. Time was encoded by population dynamics of striatal neurons in a manner that predicted duration judgments, striatal muscimol infusions degraded performance, and cooling striatal tissue led to underestimation of interval duration, suggesting that striatal dynamics underlie animals’ timing behavior. Using fiber photometry, we found that higher/lower dopaminergic activity predicted under/overestimation of interval duration. Surprisingly, optogenetic activation/suppression caused under-/over-estimation of interval duration. These data suggest that interactions between dopamine neurons and striatal networks can cause variability in timing, with broad implications for reinforcement based decision-making.

3. An Amygdala-Striatal Network for Temporal Expectation of an Aversive Stimulus (V. Doyère)

In Pavlovian aversive conditioning, the subject not only learns an association between a neutral conditioned stimulus (CS) and an aversive unconditioned stimulus (US), but also that the CS predicts the time of arrival of the US. I will present a series of experiments using local field potential recordings in awake rats during auditory aversive conditioning, which show dynamical neural activities between the dorsomedial striatum and the basolateral amygdala as a correlate of CS-US interval processing. Results show that these structures belong to a common functional network during the temporal expectancy of the US arrival, and that adaptation to new temporal contingencies involves plasticity in the striatum under amygdala control. I will discuss how this network may participate in the encoding of the CS-US interval over the course of learning, in parallel to the development of behavioral temporal control, at least in Pavlovian aversive conditioning in a supra-seconds range.

4. Timing Is an Intrinsic Computation of Neural Circuits (D. V. Buonomano)

Time is at the core of the brain’s main functions: 1) the brain stores information about the past in order to allow animals to predict and prepare for the future—the degree to which animals succeed in predicting the future translates into the evolutionary currency of survival and reproduction; 2) The brain tells time across scales ranging over 12 orders of magnitude because timing is not only critical for decoding sensory information and producing motor responses, but for prediction—the circadian clock is in sense a prediction device, it predicts when the sun will rise. Because time is so fundamental to brain function we argue that timing is highly distributed process, one that is embedded at the deepest levels into the brain’s hardware. In other words, most neural circuits are able to tell time and process temporal information in one form or another.


Gibbon, J. (1977). Scalar expectancy theory and Weber’s law in animal timing. Psychol. Rev., 84, 279–325.

Ivry, R. B. & Schlerf, J. E. (2008). Dedicated and intrinsic models of time perception. Trends Cogn. Sci., 12, 273–280.

Merchant, H., Harrington, D. L., & Meck, W.H. (2013). Neural basis of the perception and estimation of time. Annu. Rev. Neurosci., 36, 313–336.

Listen to Your Heart: Our Inner Perception and Experience of Time

Nicola Cellini 1 , Giovanna Mioni 1 , Marc Wittmann 2 , Alexandre C. Fernandes 3 , Teresa Garcia-Marques 3 and Olga Pollatos 4

1Department of General Psychology, University of Padova, Padova, Italy

2Institute for Frontier Areas in Psychology and Mental Health, Freiburg, Germany

3ISPA – Instituto Universitário William James Center of Research

4Clinical and Health Psychology, Institute of Psychology and Education, Ulm University, Germany

Symposium organizers: Nicola Cellini and Giovanna Mioni

Recent studies suggest a key role of bodily signals in the cognitive processing of time. Indeed, it has been showed that basal physiology as well as arousing situations, such as changes in body temperature, skin conductance level, muscle reactivity, and cardiac activity may affect our perception of time. For example, studies have showed how subjective experience of time is lengthened when physiological arousal increases as results of presentation of emotional stimuli, stressing situations, or mood variations. Based on these results, it has been proposed that physiological changes may not just represent the output of a possible pacemaker interfering with subjective time, but they can work as timekeepers themselves. It has been also proposed that time perception may be an embodied property of our cognition, which relies on affective and interoceptive states, that are dependent on internal bodily signals. In other words, the subjective temporal experience of external events relies on physiological rhythms, which are then integrated by cortical networks responsible for integrating bodily signals (Wittmann, 2013). This suggests that the experience of time is the result of a combination of visceral integration (i.e., interoceptive awareness) and autonomic (sympathetic/parasympathetic) control and it is based on the temporal integration of afferent signals from the body itself (Craig, 2002). Although these fascinating models aim to integrate classical cognitive models with physiological indices, the studies supporting them are showing inconsistent results, mainly due to experimental and methodological challenges. In this symposium, we aim to describe the state of art of the field and we aim to open a discussion about how body and mind are working together to keep the clock ticking and to track the passage of time, and to try to overcome the challenge this field is facing.


Heart rate variability, interoceptive processes, embodiment, electromyography, physiology

1. Waiting through Time: How the Bodily Self Shapes the Experience of Time (M. Wittmann)

Based on conceptual considerations in neuroscience and phenomenology intertwined affective and interoceptive states create the experience of time. Subjective time emerges especially in waiting situations with longer duration. A series of studies with time intervals in the range of several minutes were conducted and which were filled with different situational contexts: (1) individuals with higher scores in impulsivity and present-fatalistic time orientation relatively overestimate duration; (2) watching a staged professional dance, spectators who pay more attention to the own body signals feel that time is relatively slowed down; (3) after body-centered meditation, individuals experience waiting in a more relaxed way, time to pass faster, and they are less future oriented; (4) effects on the timing of perceived stimuli right after meditation are influenced by heart rate variability recorded during meditation. Subjective time thereafter is modulated through time- related personality factors and variations in states of the embodied self during waiting.

2. Cardiac Activity Modulates Temporal Perception (N. Cellini and G. Mioni)

Recent studies have shown that physiological signals (i.e., skin conductance) may predict and modulate our precision in temporal tasks. In two studies, we aimed to explore the relationship between temporal abilities and autonomic activity at rest and after a stressful situation. In the first study participants were asked to perform a time-bisection and two finger tapping tasks. In the second study, students were tested in the same time bisection task after either performing a stressing attentional task or the non-stressing version of the task. Prior to these tasks, we measured several electrophysiological indices at rests and during performance. Our results showed that increased heart rate variability at baseline was associated with higher temporal accuracy. Also, after the stressful situation, which increases heart rate, participants showed a more accurate and less variable temporal performance. Our results are consistent with the idea that bodily signals may shape our perception of time.

3. Spontaneous Facial Muscle Activity Predicts Duration Estimation (A. C. Fernandes and T. Garcia-Marques)

In this paper we address the hypothesis that temporal dynamics of spontaneous muscle activity may index the subjective representation of objective duration. Neural motor system (SMA) was already suggested to be involved in building-up (ramping activity) a representation of duration in explicit perceptual timing tasks. Also sustained attention, critical to time perception, was suggested to be associated with electromyographic-gradients activation varying proportionally to non-motor task’s duration. We test our hypothesis on a perceptual timing task where familiar and non-familiar targets were presented in different durations, monitoring two facial-muscles’ activity: the zygomaticus-major (supposedly indexing familiarity positivity) and the corrugator-supercilii (supposedly indexing attention). Data corroborate our hypothesis showing spontaneous electromyographic activity of each muscle to reflect different timing perception processes. Whereas the ramping corrugator activity (its duration and amplitude) reflected objective-duration as being a significant predictor of subjective duration estimates, the zygomatic overall activity predicted familiarity (a non-temporal factor) subjective duration bias.

4. Perception of Time and Body Awareness (O. Pollatos)

Internal signals like one’s heartbeats are centrally processed via specific pathways and both their neural representations as well as their conscious perception (interoception) provide key information for many cognitive processes. Recent research suggest that bodily systems like the cardiovascular system might be specifically associated with this ability, while few studies have assessed this relationship in more detail and therefore should be presented here with the focus on interoceptive signal processing. Two main tasks (retrospective time estimation, time interval reproduction) were used, while either interoceptive focus was manipulated or heartbeat information and interoceptive accuracy were assessed. Main result suggests that heart cycle and time reproduction measures responses are associated for certain time interval lengths. Information obtained from the cardiac cycle is relevant for the encoding and reproduction of time in the time span of 2 to 25 seconds. Sympathovagal tone as well as interoceptive processes mediate the accuracy of time estimation. Retrospective temporal distortions are directly influenced by attention to bodily responses. Sympathetic nervous system activation affecting memory build-up might be the decisive factor influencing retrospective time judgments, highlighting the relevance of interoception for the effects of emotional states on subjective time experience.


Craig, A. D. (2002). How do you feel? Interoception: The sense of the physiological condition of the body. Nat. Rev. Neurosci., 3, 655–666.

Wittmann, M. (2013). The inner sense of time: How the brain creates a representation of duration. Nat. Rev. Neurosci., 14, 217–223.

Circadian Rhythms in Health and Disease

Etienne Challet 1 , Claude Gronfier 2 , Fabien Pifferi 3 , and Valérie Simonneaux 3

1Regulation of Circadian Clocks Team, Institute of Cellular and Integrative Neurosciences, CNRS and University of Strasbourg, France

2Inserm U1208, Stem Cell and Brain Research Institute, Univ Lyon, Bron, France

3UMR CNRS MNHN MECADEV - Adaptative Mecanisms and Evolution 1, Brunoy, France

Symposium organizer: Valérie Simonneaux

All biological functions display daily rhythms, which are synchronized to the environment and among each other by a network of primary and secondary biological clocks. This symposium will discuss the mechanisms underlying the daily regulation of essential functions like sleep, locomotor activity and feeding through the use of unconventional animal models. It will also report how desychronization of daily rhythms has negative impacts on health and ageing processes.

1. Circadian Rhythms and Metabolism (E. Challet)

Most biological functions, including metabolism, display circadian rhythms. This temporal order is controlled by a network of circadian clocks, comprising a master clock in the hypothathamic suprachiasmatic nuclei, mainly reset by ambient light, and secondary clocks in the brain and peripheral tissues, synchronized by the master clock and mealtime. Calorie restriction and diet-induced obesity differentially modify the suprachiasmatic clockwork. Metabolic pathologies are frequently associated with circadian disorders. Conversely, epidemiological studies indicate that chronic shiftwork is highly correlated with increased incidence of type 2 diabetes, obesity, and cardiovascular disease. Furthermore, circadian desynchronization induced by chronic jetlag in rodents negatively impacts metabolism by triggering increased adiposity and impaired glucose tolerance. Chronic jetlag also leads to a shortening of telomeres, a biological marker of cellular aging. In conclusion; there are reciprocal interactions between metabolism and circadian clocks. Metabolic diseases induce circadian disturbances while circadian desynchronization is a potential cause for the metabolic syndrome.

2. Sleep and Biological Rhythms: It’s a Matter of Time! (C. Gronfier)

Sleep remains a mystery for most of us. It is critical for life since it is shared across all animal species: fishes, birds, mammals, and even worms and insects. The structure of sleep, its duration and its timing during the day are different across species, and even within species. Some are diurnal and sleep at night, others are nocturnal or crepuscular. Some are long sleepers, others are short sleepers. The same is found in humans. Within a same individual, sleep changes with age, from birth to death, in its structure, duration, and timing. The circadian biological timing system is the central master clock driving these phenomena. Sleep is a matter of time. The biological clock is the time keeper that orchestrates the physiological functions involved across the body. Sometimes, either due to internal or external factors, the symphony becomes cacophony, sleep is disturbed, the clock desynchronized, and health is compromised.

3. Is Biological Clock at the Core of Aging Process? Contributions from a Photoperiodic Non-Human Primate (F. Pifferi)

The ability of organisms to adapt to their environment during aging is altered. Age-related disorders in Human include disturbances of biological rhythms, especially sleep-wake rhythms alterations, and perturbations of body temperature and hormone secretion. The alteration of biological rhythms with age leads to major health consequences, particularly due to the alteration of sleep-wake rhythms that causes a strong alteration of the general condition. The study of these changes is therefore a major health issue and requires the use of appropriate animal model such as the grey mouse lemur (Microcebus murinus), a small Malagasy primate with very pronounced biological rhythms. Studies from this species brought useful information on the role of biological rhythms (both circadian and seasonal) and biological clock in health and longevity. On the one hand, results on the relationship between aging and the clock, illustrated by rhythms alteration in elderly individuals will be presented. On the other hand, the demonstration that the clock, through the alternation of seasonal cycles, may be key process in aging will be discussed. A final illustration of the major role played by the biological clock in the aging process is provided by the circadian resonance theory, which suggests that the closer the endogenous period of an individual or species is to 24h, the longer is its longevity. First physiological exploratory data of the underlying mechanisms of this theory in mouse lemur will also be presented.

Embodied Timing: The Role of Emergent and Predictive Timing Mechanisms in the Voluntary Control of Whole Body Movements

Yvonne Delevoye-Turrell 1 , Daniel Lewkowicz 1 , Juliane J. Honisch 2 , Mark T. Elliott 3 , Nori Jacoby 4 , Alan M. Wing 5 , Pieter-Jan Maes 6 , Valerio Lorenzoni 6 , Joren Six 6 , Simone Dalla Bella 7,8,9 , Valérie Cochen de Cock 10,11 , Dobromir Dotov 7,12 , Sophie Bayard 13 , Christian Geny 11 , Petra Ihalainen 7 and Benoît Bardy 13,8

1SCALab – UMR CNRS 9193, Room. A4.319, Building A, Domaine du Pont de Bois, 59653 Villeneuve d’Ascq, France

2School of Psychology and Clinical Language Sciences, University of Reading, Reading, RG6 6AL, UK

3Institute of Digital Healthcare, University of Warwick, Coventry, CV4 7AL, UK

4Department of Brain and Cognitive Sciences, Massachusetts Institute of Technology, Cambridge, Massachusetts, USA

5School of Psychology, University of Birmingham, Edgbaston, B15 2TT, UK

6IPEM, Dept. of Art, Music and Theatre Sciences, Ghent University, Ghent, Belgium

7EuroMov Laboratory, Montpellier University, Montpellier, France

8Institut Universitaire de France (IUF), Paris, France

9International Laboratory for Brain, Music, and Sound Research (BRAMS), Montreal, Canada

10Beau Soleil Clinic, Montpellier, France

11Centre Hospitalier Universitaire de Montpellier, Montpellier, France

12Centro de Ciencias de la Complejidad, Universidad Autónoma de México, Brasil

13Department of Psychology, Paul Valery University, Montpellier, France

Symposium organizer: Yvonne Delevoye-Turrell

Intentional motor behaviour requires the planning of body actions through space and time. Depending on the nature of the task, the goal can be geared towards the needs to produce actions at a specific moment in time. For example, clapping to a beat requires a person to set the focus on the time intervals between successive arm strokes to perform a given rhythmic pattern. In other cases, timing is simply an emergent property that reveals itself when producing cyclic movements through space. Limb velocities contain inherent dynamics that will pace motor actions at a given tempo. Such phenomenon can be observed in such cases as waving good-bye, knocking at the door or going out for a Sunday jog. The study of motor timing was initiated by Fraisse and collaborators back in the 60’s using a very simple finger tapping task that provided the means to reveal not only the natural spontaneous tempo of body movement but also the ability of a person to synchronise motor elements to an external metronome. Concerned by the emergent properties of motor timing, other studies developed the use of repetitive wrist/arm flexions to study the emergence of rhythmic coupling. This important literature has suggested through the years that predictive timing and emergent timing modes of motor control may in fact be related to different neurobiological mechanisms. Depending on the temporal constraints set upon the task, one or other of the timing modes may be implicated for motor planning and execution. Nevertheless, such suggestions are based solely on the use of simple one joint action patterns of an individual’s performance. The objective of the present symposium is to report studies focused on the timing properties of motor behaviour in whole body movement by exploring timing strategies within solo and group scenarios. This symposium will present four studies in which whole body 3D kinematics were used to reveal temporal patterns of motor control when interacting in an emotional contained environment. Overall, we aim to show how 3D kinematics can provide a way to confirm the complementary role of the automatic emergent and the cognitively controlled time measurements for optimized intentional motor behaviour. Data from healthy participants and patients with movement disorders will be presented.


3D kinematics, interval timing, synchrony, music, emotion, behaviour, neuropsychology, time series analysis, adult, Parkinson, 100s of ms-secs, secs-mins

1. Cue Properties Change Timing Strategies in Group Synchronisation (J. J. Honisch, M. T. Elliott, N. Jacoby, and A. M. Wing)

Research suggests that individuals optimise their timing performance by minimising variability of timing errors between external cues and their own movements. However, at the cost of increasing the timing variability of their own movements. The present study investigated whether an individual’s timing strategy changes according to the task, in a group scenario. We employed a novel paradigm that positioned six individuals to form two chains with common origin and termination on the circumference of a circle. We found that participants with access to timing cues from only one member used a strategy to minimise their asynchrony variance. In contrast, the participant at the common termination of the two chains, who was required to integrate timing cues from two members, used a strategy that minimised movement variability. We conclude that humans are able to flexibly switch timekeeping strategies to maintain task demands and thus optimise the temporal performance of their movements.

2. When your Movements Betray your Feelings: Reading Emotional States Through the Analysis of Spontaneous 3D Dynamics of Whole Body Motion (Y. Delevoye-Turrell and D. Lewkowicz)

Body language is a particularly challenging area of research because of the complexity of biological motion. Indeed, the human body is a tool for intentional actions (e.g., walking to close the door) but also the medium for emotional expression (e.g., running under fear). Recent 3D kinematic studies have shown that social intention modulates the timing of 3D body kinematics. In the present study, while controlling for intention, we aimed to reveal the effects of emotional states on the timing properties of 3D body kinematics when walking at a preferred and natural pace. We will report results indicating how timing properties are affected by the 5 basic emotions. Furthermore, we will suggest that emotion valence impacts specifically movement fluency and cross-body coordination, factors that could be implemented in algorithms for the automatic detection of emotional states across a multitude of real-life situations.

3. Enhancing Spontaneous Synchronization of Cyclists’ Pedal Cadence to External Music, Through Sonification of Motor Rhythms (P.-J. Maes, V. Lorenzoni and J. Six)

It is well known that music stimulates human movement. Thereby, people tend to entrain their movements to musical patterns, such as the musical beat (‘tactus’). This principle of ‘spontaneous synchronization’ has been applied in strategies to (unconsciously) adapt the tempo of rhythmical movement activities. However, this tendency for spontaneous synchronization does not seem to be consistent across people and types of rhythmical activity. In the present study, we test novel strategies to reinforce cyclists’ spontaneous tendency to synchronize their pedal cadence (motor rhythm) to an external beat (musical rhythm), and consequently, to adapt their pedal cadence in predictive manners. These strategies rely on the sonification of cyclists’ motor rhythms, allowing them to have an active contribution to the external music. Accordingly, we hypothesize that cyclists will be “attracted” to synchronized auditory-motor states as only these states lead to rewarding (synergistic) musical outcomes.

4. Predicting Individual Response to Rhythmic Auditory Cueing in Parkinson’s Disease (S. Dalla Bella, V. Cochen de Cock, D. Dotov, S. Bayard, C. Geny, P. Ihalainen and B. Bardy)

Gait dysfunctions in Parkinson’s disease (PD) can be partly relieved by rhythmic auditory cueing. This consists in asking patients to walk with a rhythmic auditory stimulus like a simple metronome or music. This beneficial effect on gait is visible immediately during the stimulation in terms of increased speed and stride length. In spite of the beneficial effects of cueing at the group level, however, individual patients tend to react differently to auditory cues. In this study we measured the response of 39 patients with PD and 39 controls to auditory rhythmic cues (metronomes and music), and their rhythm perception and synchronization abilities. Gait spatio-temporal parameters were recorded while the participants walked together with rhythmic cues. Only some patients clearly benefitted from cueing while others exhibited deleterious effects of the stimulation. This effect depends on patients’ rhythmic abilities. These findings pave the way to devising individualized gait rehabilitation protocols.

Timing and Time Perception in Children

Sylvie Droit-Volet 1 , Florie Monier 1 , Jennifer T. Coull 2 and Laurence Casini 3

1Université Clermont Auvergne, Laboratoire de Psychologie Sociale et Cognitive, CNRS, Clermont-Ferrand, France

2Laboratoire de Neurosciences Cognitives (LNC), Aix-Marseille Université, CNRS (UMR7291), Marseille, France

3Aix-Marseille Université, CNRS, Marseille, France

Symposium organizer: Sylvie Droit-Volet

A symposium focused on timing and time perception in children.


Babies, children, timing, cognition, disorder

1. Different Developmental Trajectories for Different Time Judgments (S. Droit-Volet)

There is not one but different types of judgment of time. Developmental studies in children aged from 5 to 9 years and adults have shown that developmental trajectories change between implicit and explicit timing. In implicit timing task, temporal variability remains constant across different ages but changes with age in explicit timing task, although the rhythm of development depends on the task used. Moreover, for explicit judgments of time, children are more prone to time distortion, being highly sensitive to contextual effects. These age-related changes in timing are significantly linked to the development of general cognitive capacities, in terms of working memory and attention. We will, therefore, discuss the development as an enrichment of separate forms of representation of time that are increasingly integrated though the development of awareness of the passage of time.

2. Action Helps Young Children Maintain a Robust Representation of Time in Memory (F. Monier, S. Droit-Volet and J. T. Coull)

Studies have shown that motor reproduction of a rhythm improves with age, and is linked to cognitive and motor development. However no study has yet examined whether developmental progress influences the motor precision of rhythm production, the temporal encoding of the inter-stimuli interval (ISI), or some combination of the two. 5 and 8 year old children watched a rhythmic sequence of dots (800 ms ISI). Within each age group, half of the participants tapped coincidently with the onset of the dots (visuo-motor learning), whereas the other half simply looked at these dots (visual learning). Upon extinction of the visual input, children reproduced this rhythm three times in a row, without re- presentation of the visual rhythm. Children’s individual motor and memory capacities were assessed with neuropsychological tests. Our results showed that although 5 year olds were globally less accurate than 8 year olds, all children accurately reproduced the rhythm at least during the first reproduction trial. During the second and the third trials, on the other hand, tapping rate accelerated, especially in the younger children. Moreover, performance in the second and third trials was significantly less accurate for 5 year olds in the visual learning condition than for those in the visuo-motor one. Visuo-motor learning thus improved the ability to maintain the representation of the rhythm over the three trials. Correlation with neuropsychological measures confirmed this interpretation, revealing that temporal performance in the first reproduction trial was significantly linked to fine motor abilities, whereas temporal performance in the second and third trials was related to short-term and working memory capacity. Consequently, it seems that motor synchronization not only improves the motor component of rhythm reproduction, but also reinforces representation of the learned interval in memory.

3. Time, Speech, and Dyslexia (L. Casini)

The prevailing view concerning the cause of dyslexia in children points to phonological processing difficulties. However, it is still unclear whether the phonological deficit is the primary cause of dyslexia or whether it is secondary to impairments in the processing of more basic acoustic parameters of the speech signal. A recent theory of dyslexia, the temporal sampling theory, has put back temporal processing in the center of attention (Goswami, 2011). It assumes that deficits in temporal sampling could explain abnormal phonological development in children with dyslexia. In this talk, I will present some studies investigating the link between time, speech and dyslexia, and I will attempt to shed new light on temporal processing deficits in dyslexia by using the theoretical and methodological tools of time perception.


Goswami, U. (2011). A temporal sampling framework for developmental dyslexia. Trends Cogn. Sci., 15, 3–10.


(in order of presentation)

Implicit Temporal Predictability Improves Auditory Pitch Discrimination Sensitivity

Sophie K. Herbst 1,2 , Virginie van Wassenhove 2,3,4 and Jonas Obleser 1

1Department of Psychology, University of Luebeck, Germany

2CEA, NeuroSpin, France

3INSERM, U992, Cognitive Neuroimaging Unit, Université Paris-Sud, France

4Université Paris-Saclay, F-Gif/Yvette, France

The human brain automatically extracts temporal contingencies from the environment, that is, rhythms or indicative durations, which allow the prediction of future events. While numerous studies have targeted rhythmic temporal contingencies, others have assessed temporal predictability by manipulating the time interval between cue and target in a trial-by- trial design, i.e., the foreperiod. Commonly, the use of temporal predictability in foreperiod paradigms has been promoted by using a few discrete durations, or by making participants aware of the temporal contingencies with explicit temporal cueing. Here, we aimed at testing the use of temporal predictability in a more naturalistic scenario, in which probabilistic variations of foreperiods were introduced implicitly, that is, without giving explicit temporal cues or informing participants of the relevance of timing. To this end, we varied foreperiods in an auditory pitch discrimination task in three independent EEG and psychophysical experiments. Unanimously, all three studies provide evidence that implicit probabilistic temporal predictability is used: Predictability sped up response times, or, depending on the experimental paradigm, improved pitch discrimination sensitivity. Concerning the neural signatures of temporal predictability, comparing predictive with non- predictive experimental conditions revealed a reduced cue-related evoked P2 response and an increased central alpha (7–12 Hz) power during predictive foreperiods. We also expected to find entrainment of the slow frequency at the stimulation rate, so that the target presented at the predicted time point would fall at the preferred phase of the entrained oscillation. However, this was not the case in the first two studies, suggesting that this mechanism might be contingent on more explicit forms of temporal predictability or more rhythmic scenarios. The third study was designed for a model based EEG-approach, to potentially reveal subtle effects of temporal predictability with the phase of neural oscillations, or, allowing us to underline the previous conclusions that these effects are contingent on more rhythmic paradigms.


Prediction, behaviour, EEG, adult, 100s of ms-secs

Saccadic Inhibition as an Index of Anticipation in a Discrimination Task

Roy Amit 1 , Dekel Abeles 2 , Marisa Carrasco 3 and Shlomit Yuval-Greenberg 1,2

1Sagol School of Neuroscience, Tel Aviv University, Israel

2School of Psychological Sciences, Tel Aviv University, Israel

3Department of Psychology and Center for Neural Science, New York University, USA

Predicting the timing of upcoming events improves resource allocation and action preparation. Temporal anticipation is most commonly estimated using behavioral measures: when visual stimulus onset can be predicted, its perception is enhanced, as indicated by increased accuracy-rates and decreased reaction times in detection and discrimination tasks. However, these measures are indirect as they evaluate temporal anticipation retrospectively rather than index the process in real- time. A recent study (Dankner et al., in press) has shown that temporal predictions can be assessed directly by measuring pre-stimulus suppression of saccades in a detection task. Here we investigated whether the inhibition of fixational saccades can serve as an index of temporal anticipation in a discrimination task and compare to behavioral indices of temporal sensitivity. In each trial, observers were requested to discriminate the orientation of a central slightly tilted grating stimulus. This stimulus was preceded, with a varying stimulus-onset-asynchrony (SOA), by a cue, which was either informative or non-informative regarding the timing of the target. Our results showed that fixational saccades were suppressed before a predictable, but not before an unpredictable stimulus. This effect decreases linearly as SOA duration increases and is more robust than accuracy or RT measures. Additionally, a Bayesian classifier was trained to predict the stimulus’ class (predictable/unpredictable) based on the single-trial oculomotor events preceding their onset with 63.8% mean accuracy across subjects. We conclude that the oculomotor system is sensitive to anticipation, and that saccadic inhibition can be used as an index for temporal anticipation, even when little or no effect is seen in other measures.


Prediction, duration, behavior, electrophysiology, adult, 100 s of ms-secs


Dankner, Y., Shalev, L., Carrasco, M., & Yuval-Greenberg, S. (in press). Pre-stimulus inhibition of saccades in adults with and without ADHD as an index for temporal expectations. Psychol. Sci.

Neurological Evidence of a Dual Origin of the Foreperiod Effect

Bertrand Degos 1,2 , Ilhame Ameqrane 3,4 , Sophie Rivaud-Péchoux 4 , Pierre Pouget 4 and Marcus Missal 3,4

1Neurology department, Parkinson’s disease expert centre, Salpêtrière Hospital AP-HP, Paris, France

2Center for Interdisciplinary Research in Biology, Collège de France, INSERM U1050, CNRS UMR7241, Labex Memolife, Paris, France

3Institute of Neuroscience (IONS), Cognition and Systems (COSY), Université catholique de Louvain, Brussels, Belgium

4Sorbonne Universités, UPMC Univ Paris 06, Inserm U1127, CNRS UMR 7225, UM 75, ICM, F-75013 Paris, France

We investigated the foreperiod effect on the latency of saccadic eye movements in Parkinson’s disease (PD) patients. Patients suffered either of the idiopathic (referred to as ‘iPD’) or of a rare genetic variant of the disease (referred to as ‘Parkin’). In controls and in all patients saccadic latencies were longer for short foreperiods and decreased for longer ones, showing an oculomotor instantiation of the variable foreperiod effect. However, in controls and Parkin patients, the latency of saccades was also influenced by the duration of the previously experienced foreperiod. On average, this temporal short-term memory effect was not observed in iPD patients, either ON or OFF L-Dopa therapy. We suggest that the influence of the hazard rate on movement preparation is not affected in PD and is not dopaminergic-dependent. However, the absence of a contribution of short-term temporal memory to the FP effect in iPD patients clearly shows the dual origin of the foreperiod effect.

Prospects of a Multiple Trace Theory of Temporal Preparation

Sander Los, Wouter Kruijne and Martijn Meeter

Vrije Universiteit Amsterdam, The Netherlands

In warned reaction time tasks, the warning stimulus (S1) initiates a process of temporal preparation, which promotes a speeded response to the impending target stimulus (S2). Classic theories assume that temporal preparation is under strong voluntary control, informed by the distribution of S1-S2 intervals. However, our recently developed multiple trace theory of temporal preparation (MTP) offers a more mechanistic insight into preparation, based on simple associative learning rules. Here, we present recent experiments where different groups of participants experienced either the exponential or antiexponential distribution of S1–S2 intervals during an acquisition phase. This was followed by a test phase where all participants received, after explicit instruction, the uniform distribution. In this phase we found highly persistent transfer effects from the acquisition phase. These long-term learning effects are hard to reconcile with classic theories of temporal preparation and provide strong support for MTP.

Task-Oriented Optimal Inference in Interval Timing

Zhenguang Cai 1 and Maarten Speekenbrink 2

1School of Psychology, University of East Anglia, UK

2Department of Experimental Psychology, University College London, UK

Resent research on interval timing has taken a Bayesian inference approach, according to which people make an optimal inference about the magnitude of a duration based on their noisy perception and their prior temporal experience (Cicchini et al., 2012; Jazayeri & Shadlen, 2010). However, the specific mechanism behind Bayesian inference in timing is yet to be explored. We propose that people make task-oriented optimal inferences. In time reproduction, for instance, people infer how much time to be reproduced (rather than how much time there is in the stimulus) based on both the perceived stimulus duration and their prior reproduction experience (rather than their prior perception experience). This proposal thus contrasts with existing Bayesian timing accounts where the inference is an updated perceived duration in light of prior perception experience (Petzschner et al., 2015). To test the proposal, we conducted three set of experiments using time reproduction. In particular, we make use of the range bias (a duration is reproduced as longer if it follows longer than shorter prior durations), which has been shown to be driven by the use of prior experience in inferences (Jazayeri & Shadlen, 2010). The first two experiments show that it is the reproduction rather than perception experience that constitutes the prior and drives the range bias in time reproduction. The second set of two experiments interleave motor vs. auditory reproduction in the same session and show that people resort to task-specific priors (based on prior motor or auditory reproductions) such that the different task-specific priors leads to a range bias. Finally, two experiments further show that reproduction is influenced by the reproduced but not perceived duration in the preceding trial, suggesting a trial-by-trial updating of the prior based on reproduction experience. All these results support the task- oriented optimal inference account.


Duration, behavior, adult, 100s of ms-secs, Bayesian inference, range bias


Cicchini, G. M., Arrighi, R., Cecchetti, L., Giusti, M., & Burr, D. C. (2012). Optimal encoding of interval timing in expert percussionists. J. Neurosci., 32, 1056–1060.

Jazayeri, M., & Shadlen, M. N. (2010). Temporal context calibrates interval timing. Nat. Neurosci., 13, 1020–1026.

Petzschner, F. H., Glasauer, S., & Stephan, K. E. (2015). A Bayesian perspective on magnitude estimation. Trends Cogn. Sci., 19, 285–293.

Human Perceived Timing Follows Principles of Bayesian Inference

Darren Rhodes 1,2 and Massimiliano Di Luca 1

1Centre for Computational Neuroscience and Cognitive Robotics, School of Psychology, University of Birmingham, Edgbaston, Birmingham, UK

2Sackler Centre for Consciousness Science, School of Engineering & Informatics, University of Sussex, Brighton, UK

The timing of events is fundamental for everyday life, yet how humans perceive time remains unknown. Here we present a Bayesian inference model for the perceived timing of stimuli that combines noisy measurements with dynamic prior expectations. Quantitative validation of the model is difficult, as the precise noise characteristics of the priors and likelihoods are unknown, yet here we are able to infer such characteristics from psychophysical data. The model predicts that stimuli presented earlier or later than expected are reported closer to the expected timing, an effect that should increase with unreliable information. We presented sequences of regularly timed auditory stimuli with alternating intensities and asked participants to report whether the final stimulus appeared before or after a visual probe. The final stimulus was either of high or low intensity so to vary the reliability of estimating perceived timing. In line with the predictions of the model, we find that perceived simultaneity of stimuli is drawn towards the expected time point and the effect is larger for low- intensity stimuli. The novel account we propose is a viable alternative to existing models of perceived timing. The framework is general and can be applied to a wide spectrum of investigation in the psychology and neuroscience of time perception.


Bayesian inference, timing, rhythm, uncertainty, perception, perceived timing

How Pain Affects Time Estimation. A Physiological Study

Andrea Piovesan, Laura Mirams, Helen Poole, David Moore, Michael Richter and Ruth Ogden

Liverpool John Moores University, UK

Time experience is known to be distorted by emotional stimuli (Droit-Volet et al., 2013). A proposed explanation for this is that emotional stimuli affect arousal. Indeed it is often suggested that arousal moderates time experience (Gil & Droit-Volet, 2012). However, the relationship between physiological arousal and perceived duration is often assumed rather than tested. In a series of studies, the relationship between physiological arousal and temporal perception was directly tested. Healthy adults were asked to judge either 1) the duration of a painful electric shock, or, 2) the duration of neutral visual stimulus whilst in a prolonged state of pain. In both cases, participants experienced different subjective pain intensities (previously rated using a Numerical Rating Scale) to induce different arousal levels. Measures of skin conductance level and high-frequency heat-rate variability were also recorded as indicators of the sympathetic and parasympathetic arousal, respectively. Results indicate that sympathetic arousal is related to temporal judgments. However, this was true only when the arousing stimulus is the to-be-timed stimulus (electric shock). When arousal originates from a secondary, non-timed source, arousal may have the capacity to distract from the process of timing.


Duration, behavior, physiology, adult, 100s of ms-secs


Droit-Volet, S., Fayolle, S., Lamotte, M., & Gil, S. (2013). Time, emotion and the embodiment of timing. Timing Time Percept., 1, 99–126.

Gil, S., & Droit-Volet, S. (2012). Emotional time distortions: The fundamental role of arousal. Cogn. Emot., 26, 847–862.

Caloric Rewards Alter Time Perception and Time-Dependent Decision Making

Bowen J. Fung 1,2 , Carsten Murawski 2 and Stefan Bode 1

1Melbourne School of Psychological Sciences, The University of Melbourne, Victoria, Australia

2Department of Finance, The University of Melbourne, Victoria, Australia

Time is a critical component of many important decision-making processes. One example of this is in foraging behaviour, which requires a trade-off between the value of potential future rewards and the opportunity cost of time. However, not much is known about the subjective representation of time in these decision-making processes. Furthermore, the importance of accurate time perception in decision making seems at odds with our general experience of time, which is highly labile to both external and internal factors. Some researchers have previously suggested that distortions in our sense of time allow us to make better decisions by flexibly adapting to the current ecological context. For example, individuals have been shown to underestimate the duration of images of food, and this underestimation is larger for subjects who are more hungry (Gable & Poole, 2012). One potential explanation is that time might be underestimated in order to minimise the perceived cost of acquiring food (if durations are under- estimated, the pursuit of reward might be prolonged). Here I report two studies that explored the psychophysiological relationships between biologically relevant rewards, time perception, and ecological, time-dependent decision making. In the first study, we recruited 125 participants and assessed their performance on a novel variant of a duration production paradigm. While they performed this task, participants received different magnitudes of rewards on a trial-by-trial basis. We tested four different types of liquid primary rewards (fruit juice, water, aspartame, and maltodextrin) and a secondary reward (money). We found that consumption of the primary rewards containing calories lead to systematic overproductions of intervals (from 2-5 seconds). This pattern of results suggested that biologically relevant rewards were able to alter time perception (Fung et al., 2017b). In the second study, we recruited 50 participants to investigate whether caloric primary rewards were also able to bias time-dependent decision making. Participants fasted for four hours, and then completed a task similar to a patch-leaving foraging paradigm, incentivised with monetary rewards. Participants who consumed a caloric drink in between blocks gave up waiting for rewards significantly earlier, compared to those who consumed water (i.e., participants who consumed the caloric drink were less patient.) These results suggested that the consumption of biologically relevant rewards was able to alter time-dependent decision making, despite the fact that the drink was irrelevant to the task (Fung et al., 2017a). These results support the idea that time perception can be affected by an individual’s homeostatic state, and further suggest that different homeostatic states can influence time-dependent decision making processes. Taken together, these experiments imply that our experience of time may be part of a psychophysiological mechanism whereby energy levels affect perceived time, which, in turn, may act to optimise ecological decision making.


Reward, decision-making, duration, behavior, adult, secs-mins


Gable, P. A., & Poole, B. D. (2012). Time flies when you’re having approach-motivated fun: Effects of motivational intensity on time perception. Psychol. Sci., 23, 879–886.

Fung, B. J., Bode, S., & Murawski, C. (2017a). High monetary reward rates and caloric rewards decrease temporal persistence. Proc. R. Soc. Lond. B Biol. Sci., 284.

Fung, B. J., Murawski, C., & Bode, S. (2017b). Caloric primary rewards systematically alter time perception. J. Exp. Psychol. Hum. Percept. Perform.

Temporal Representations in the Duration Discrimination Task

Başak Akdoğan 1 , Randy Gallistel 2 , Ben Gersten 1 , Amita Wanar 3 and Peter Balsam 3,4

1Department of Psychology, Columbia University, USA

2Department of Psychology, Rutgers University, USA

3Department of Psychology, Barnard College, Columbia University, USA

4New York State Psychiatric Institute, USA

Temporal information-processing is critical for adaptive behavior and goal-directed action. It is thus crucial to understand how the temporal distance between behaviorally relevant events is encoded. To this end, we tested mice in a duration discrimination procedure in which they learned to correctly categorize tones of different durations as short or long. To investigate the nature of temporal representations, after being trained on a pair of target intervals they transferred to conditions in which the absolute and/or relative relations of target intervals and corresponding response locations were systematically manipulated. The adjustments of choices to these changes in temporal and spatial contingencies were characterized with respect to acquisition, learning rates, temporal sensitivity, and response times associated with short and long categorizations. These indices of timing behavior collectively revealed that although subjects appear to have knowledge of absolute durations, the transfer occurred most readily when the relative mapping between durations and corresponding response locations was maintained. The findings indicate that the default representations in a duration discrimination task encode relative rather than the absolute information about time.


Duration, behavior, modeling, animal (mice), secs-mins

The Relation between Symbolic and Non-Symbolic Representations of Time

Karina Hamamouche and Sara Cordes

Boston College, USA

Studies have identified a relation between non-symbolic numerical abilities and formal mathematics (see Halberda et al., 2008), yet no work has examined whether a similar relation between informal, non-symbolic and formal, symbolic abilities exists in other quantities domains, such as time. That is, does acquiring a symbolic understanding of time (i.e., identifying time on a clock or accurately estimating how long it takes to complete everyday activities) relate to improvements on basic timing tasks? To test this hypothesis, 6–7 year old children (N = 40; Mage = 7.07 years, 22 males), who are in the process of learning symbolic time concepts in school, completed a temporal discrimination task, a temporal estimation task, and a symbolic timing assessment (which consisted of reading analog and digital clocks, estimating how many minutes it takes to complete everyday activities, etc.). Parents also completed a survey regarding their child’s symbolic timing experiences and activities. Results revealed that, when controlling for age, children’s symbolic timing performance was independently correlated with both non-symbolic discrimination performance (r = .508, p = .002) and estimation performance (r = −.421, p = .013). Further, age (β = .648, p < .001), discrimination performance (β = .313, p = .004), and estimation performance (β = −.241, p = .020) each uniquely predicted performance on the symbolic timing task when controlling for the other variables. Parent surveys, however, were unrelated to any of the child measures (p’s > .18). Paralleling findings in the numerical domain, this work is the first to identify a relation between symbolic and non-symbolic temporal abilities. Future work using experimental designs is necessary to determine whether this relation is causal.


Duration, behavior, children, 100s ms-secs, secs-mins


Halberda, J., Mazzocco, M. M., & Feigenson, L. (2008). Individual differences in non- verbal number acuity correlate with maths achievement. Nature, 455, 665–668.

Modality Specific Rate Aftereffects: Evidence towards Distributed Timing Mechanisms

Aysha Motala and David Whitaker

School of Optometry and Vision Sciences, Cardiff University, Cardiff, UK

Time perception is important for a number of different tasks, some of which include understanding speech and appreciating music. However, it remains to be elucidated whether sensory time perception occurs due to a central timing component overlooking each sense or rather, that distributed mechanisms exist each specifically dedicated to each sense, and operating in a largely independent manner. To this end, a range of unimodal and cross-modal rate adaptation experiments were conducted, in which adapting to a fast rate makes a moderate rate feel slow, and adapting to a slow rate, makes the same moderate rate feel fast. Rate perception was quantified by a method of temporal reproduction across the visual, auditory and tactile senses on three observers. Repulsive rate aftereffects were observed across all unimodal conditions however, no such effects were observed for any cross-modal pairings. A similar pattern of results was found in subsequent experiments when subjects were unaware of which sensory modality would form the test stimulus. We use the current findings to suggest that sensory timing abilities are open to change, but crucially, that this change is modality-specific and finally, that our data supports the distributed theory of timing..


Rhythms, multisensory, behavior, 10s-100s of ms, 100s of ms-secs

Observers Adapt to the Physical, not the Perceived Duration of Visual Events

Jim Maarseveen 1 , Chris L.E. Paffen 1 , Frans A. J. Verstraten 1,2 and Hinze Hogendoorn 1

1Department of Experimental Psychology, Helmholtz Institute, Utrecht University, Utrecht, The Netherlands

2School of Psychology, Faculty of Science, University of Sydney, Sydney, Australia

Recent evidence suggests that duration selective channels underlie the encoding of duration (Heron et al., 2012). Here, we investigated whether these duration selective channels encode the time between the onset and offset of a stimulus, or the perceived duration of a stimulus. To this end, we adapted participants to stimuli that differ in their physical and perceived duration. More specifically, we used the Temporal Frequency-Induced Time Dilation illusion in which the perceived duration of stimuli increases with increasing temporal frequency content, without changing the perceived onset and offset of those stimuli (Kanai et al., 2006; Kaneko & Murakami, 2009). Participants adapted to repetitions of a rotating radial grating (illusion-inducing stimulus); a static radial grating matched to the physical duration of the illusion- inducing stimulus (physically matched stimulus); or a static radial grating matched to the perceived duration of the illusion-inducing stimulus (perceptually matched stimulus). We then measured the resulting duration after-effect using a cross-modal duration judgment task in which participants compared the duration of an auditory reference to that of a static visual test stimulus. We found that participants adapted to the physical, and not the perceived duration of our illusion-inducing stimulus: the duration after-effect for the illusion-inducing stimulus did not differ from the after- effect for the physically matched stimulus; but did differ from the after-effect for the perceptually matched stimulus. This result demonstrates that duration encoding is based on the time difference between stimulus onset and offset and does not necessarily reflect the perceived duration of a stimulus.


Duration, behavior, adult, 10s to 100s of ms


Heron, J., Aaen-Stockdale, C., Hotchkiss, J., Roach, N. W., McGraw, P. V., & Whitaker, D. (2012). Duration channels mediate human time perception. Proceedings. Biological Sciences / the Royal Society, 279, 690–698.

Kanai, R., Paffen, C. L., Hogendoorn, H., & Verstraten, F. A. (2006). Time dilation in dynamic visual display. J. Vis., 6, 1421–1430.

Kaneko, S., & Murakami, I. (2009). Perceived duration of visual motion increases with speed. J. Vis., 9, 14.

Rapid Recalibration to Audiovisual Asynchronies Occurs Unconsciously

Erik Van der Burg 1,2 , David Alais 2 and John Cass 3

1Dept. of experimental and applied psychology, Vrije Universiteit Amsterdam, The Netherlands

2School of psychology, University of Sydney, Australia

3School of psychology, Western Sydney University, Australia

In natural scenes, audiovisual events deriving from the same source are synchronized at origin. However, from the perspective of the observer, there are likely to be significant multisensory delays due to physical and neurological differences. Fortunately, our brain appears to compensate for the resulting latency differences by rapidly adapting to the asynchronous audiovisual events (Van der Burg, Alais & Cass, 2013; Van der Burg & Goodbourn, 2015). Here, we examine whether rapid recalibration to asynchronous signals occurs unconsciously. On every trial, a brief tone pip and flash were presented across a range of stimulus onset asynchronies (SOAs). Participants were required to perform two tasks in alternating order. On adapter trials, participants judged the order of the audiovisual events. Here, audition either lead or lagged vision with a fixed SOA (150 ms). On test trials the SOA as well as the modality order varied randomly, and participants judged whether the events were synchronized or not. For test trials, we show that the point of subjective simultaneity (PSS) follows the physical rather than the perceived (reported) modality order of the preceding trial. These results suggest that rapid temporal recalibration occurs unconsciously.


Multisensory processing, synchrony judgment, recalibration, audiovisual


Van der Burg, E., Alais, D., & Cass, J. (2013). Rapid recalibration to asynchronous audiovisual stimuli. J. Neurosci., 33, 14633–14637.

Van der Burg, E., & Goodbourn, P. T. (2015). Rapid, generalized adaptation to asynchronous audiovisual speech. Proc. R. Soc. Lond., B, 282, e20143083.

Time Order is a Psychological Bias

Laetitia Grabot 1 , Anne Kösem 2,3 , Leila Azizi 1 and Virginie van Wassenhove 1

1Cognitive Neuroimaging Unit, Institut National de la Santé et de la Recherche Médicale, Université Paris-Saclay, NeuroSpin, Commissariat à l’Energie Atomique et aux Energies Alternatives, Université Paris-Saclay, France

2Radboud University, Donders Institute for Brain, Cognition and Behaviour, Nijmegen, The Netherlands

3Neurobiology of language, Max Planck Institute for Psycholinguistics, Nijmegen, The Netherlands

Inter-individual variability in temporal order perception has rarely been tackled empirically although readily observed in the literature. Classically, temporal order perception has been studied using temporal order judgments (TOJ) in which participants report the order of two stimuli presented few milliseconds apart. The temporal delay, at which participants consider the order of stimuli to be at chance is called Point of Subjective Simultaneity (PSS) and is typically averaged across participants canceling out potential individual effects. Here, we asked whether an individual’s PSS was stable over weeks, which would indicate an intrinsic or hardwire constant delay in perceptual systems. The present work aimed to test this stability and to distinguish this hypothesized intrinsic constant from an attentional effect, considering that temporal order perception is known to be modulated by attentional fluctuations. We designed a longitudinal psychophysics study using auditory, visual, and audiovisual TOJs. The experimental design also comprised unisensory (vision or audition attended) and divided (audition and vision attended) attentional conditions. A standard measure of PSS was extracted from the divided attentional condition, and a measure of PSS free of any attentional biases was computed from the unisensory attentional conditions. Our results show that individual standard PSS are stable over months, strengthening the hypothesis that the PSS is an individual marker of temporal perception and thus a defining feature of an individual’s core biases. Attention could partially, but not fully, compensate for this bias. We measured the neural oscillatory activity with magnetoencephalography during a TOJ task and this study revealed that perception of temporal order cannot be predicted if the individual biases are not taken into account. These results shed a new light in the debated issue of what is actually measured in a TOJ and stressed the importance to tackle individual differences.


Order, multisensory, behavior, MEG, adult, 10s-100s of ms

Reframing Variability in Auditory, Visual, and Audiovisual Timing Tasks: From Nuisance to an Aid to Understand Complex Systems Dynamics

Lars T. Boenke 1,2 , Richard Höchenberger 3 , David Alais 2 and Frank W. Ohl 1,4,5

1Leibniz Institute for Neurobiology, Magdeburg, Germany

2University of Sydney, Sydney, Australia

3German Institute of Human Nutrition Potsdam-Rehbrücke, Germany

4Otto-von-Guericke Universität Magdeburg, Magdeburg, Germany

5Center for Behavioral Brain Sciences (CBBC), Magdeburg, Germany

In modern behavioral science, audiovisual timing is regarded as the oldest example of unexplained variability across studies and across individuals. While conclusions are in general average-based and partly contradictive, variability is condoned as nuisance. While both, synchrony- (SJs) and temporal-order-judgments (TOJs), have been employed in tasks measuring the point of subjective simultaneity (PSS), the latter is found to show higher inter-individual variability than the former. This difference lacks explanation. A fundamental difference between both tasks is, that while SJs can be achieved by focusing only on the temporal relationship of stimuli, TOJs require focusing on an additional stimulus dimension (e.g., location, etc.). Processing additional stimulus dimensions need in general additional neural networks. We hypothesized that the combination of a larger number of different networks during a decision under uncertainty yield a larger number of possible states being randomly realized in each trial. Larger number of possible realizations, ceteris paribus, are accompanied by higher variability. To test for this hypothesis, we first performed a meta-study on SJ- and TOJ-tasks that have employed simple audiovisual stimuli, and plotted the average reported point PSS as a function of the number of participants employed in each study (considering each participant as one possible realization). Subsequently, we performed a within-participant study including auditory, visual, and audiovisual SJs and spatialized TOJs. Besides different task instructions, both tasks were otherwise identical. The meta-study revealed average PSS-values with a slight preference for the auditory modality (statistically indistinguishable between tasks). Moreover, the scatter of study-wise obtained PSS-values was funnel-shaped with a broader funnel for TOJs than for SJs (in line with bootstrap results in our study). Finally, simulations supported our overall results to be compatible with the notion that differing number of possible states yields differing variability irrespective of the level of observation (i.e., single-trials, between individuals or studies).

Saccadic Temporal Recalibration Leads to a Reversal of Cause and Effect

Brent Parsons, Dunia Giomo and Domenica Bueti

Scuola Internazionale Superiore di Studi Avanzati, Italy

Studies of saccadic adaptation have predominately focused on manipulations in the spatial dimension. Shifting the location of the saccade target midflight leads to changes in the motor command (e.g., saccade amplitude) and affects subsequent perceptual judgments (e.g., localization). Significant gain reduction has been reported even when the shifted target is presented at post-saccade delays of up to 400 ms (Shafer, Noto & Fuchs 2000). Recent experiments manipulating only the temporal dimension, the delay between saccade landing and target presentation, have shown changes in peak velocity of the saccade (Shadmehr et al., 2010) and the duration of the post-saccadic stimulus (Parsons & Ivry, 2014). The current study investigates whether adapting to these artificially induced delays leads to temporal recalibration between action and effect. Subjects made rapid alternating saccades with fixed delays between the end of the eye movement and the onset of the visual stimulus. After every fifth saccade, a probe stimulus was presented at one of several offsets around saccade ending. Subjects were asked if the probe onset occurred before or after their eye landed. Following adaptation to delays, subjects reported visual stimuli presented after the saccade ending as occurring before their saccade, evidence for saccadic temporal recalibration. The results provide novel insight into the mechanisms underlying perceptual stability and links saccades to the more general phenomenon of motor-sensory recalibration.


Shafer, J. L., Noto, C. T., & Fuchs, A. F. (2000). Temporal characteristics of error signals driving saccadic gain adaptation in the macaque monkey. J. Neurophysiol., 84, 88–95.

Shadmehr, R., de Xivry, J. J. O., Xu-Wilson, M., & Shih, T. Y. (2010). Temporal discounting of reward and the cost of time in motor control. J. Neurophysiol., 30, 10507–10516.

Parsons, B.D., & Ivry, R.B. (2014). Perceptual consequences of delaying the post-saccadic target. J. Vis., 14, 1228–1228.

Perceived Timing of a Visual Event Is Affected by Temporal Context

Ljubica Jovanovic and Pascal Mamassian

Laboratoire des systèmes perceptifs, Département d’études cognitives, École normale supérieure, PSL Research University, CNRS, Paris, France

Perceived timing of an event is affected by the context, as demonstrated with kappa effect (Cohen et al., 1953) or temporal ventriloquism (Fendrich & Corballis, 2001). However, it is not clear how these phenomena can be explained by models of time perception. We present here a novel paradigm that enables us to precisely study how timing of a visual event is affected by other events in temporal proximity. Participants were initially familiarized with a fixed interval duration by watching the hand of a clock rotating at a constant speed, one cycle in 2 seconds. In the rest of the experiment, the hand was no longer presented and the clock face was represented as a circle. A stimulus was then briefly flashed within the interval duration, and participants had to place a cursor on the circle at the location where the hand would have been at the time of the flash. Stimuli were discs of different colors presented at the center of the clock. In most trials (91%), two stimuli were presented and a stimulus color was shown at the end of the trial as a cue for reproduction. We varied the relative timing of target and distractors in ten steps (-300 to 300 ms). In the remaining trials, only one stimulus was shown. In additional experiments, we assessed timing of a single event and sensory - motor noise in a pointing task. Estimated timing of a target was affected by timing of distractors: targets preceded by a distractor appeared earlier than those followed by it. Performance in the single event experiment was more accurate and less variable compared to the main experiment. These results are in the line with previous findings suggesting that timing of different events are interdependent (Fendrich & Corballis, 2001) and affected both by temporal proximity as well as number of events.


Cohen, J., Hanse, C. E. M., & Sylvester, J. D. (1953). A new phenomenon in time judgment. Nature, 172, 901.

Fendrich, R., & Corballis, P. M. (2001). The temporal cross-capture of audition and vision. Percept. Psychophys., 63, 719–725.

Perceived Timing of Sensory Events Triggering Actions in Parkinson’s Disease

Yoshiko Yabe 1,2 , Penny A. MacDonald 1 and Melvyn A. Goodale 1

1The Brain and Mind Institute, The University of Western Ontario, Canada

2Research Institute, Kochi University of Technology, Japan

Recently, we showed that the perceived timing of a sensory event that triggers an action was delayed, as if it were attracted towards the action. The fact that this delay in the perceived timing occurred even when an individual cancelled an action suggests that it is the motor programming rather than the action itself that affects the perceived time (Yabe & Goodale, 2015). One can ask therefore: does this distortion in the perception of time of onset of a stimulus event disappear in Parkinson’s disease (PD) patients as a result of their impairment in preparation of actions? In this study, we examined the subjective onset of sensory events that triggered actions in PD patients. Fourteen PD patients, 15 age-matched controls and fifteen young controls were tested on a Go/No-go task, both ON and OFF dopaminergic therapy on separate days for all groups. Participants were required to fixate a clock face with a rotating second hand at the center of a computer screen. In the action condition, participants were required to blow into a microphone when they heard an auditory tone in which the pitch indicated a Go trial. They were required to cancel their actions when they heard an auditory tone in which the pitch indicated a No-go trial. In the control condition, regardless of the pitch of the tone, they were asked not to blow into the microphone. The lag of perceived onset of the tone compared to the real timing was calculated for each trial. As a result, the difference in the lags between the action and control conditions were smaller in PD patients than in both control groups. Interestingly, in PD patients in OFF medication state, the difference in the lags was significantly smaller than zero. In short, time flew backward when PD patients intended to act.


Yabe, Y., & Goodale, M. A. (2015). Time flies when we intend to act: Temporal distortion in a go/no-go task. J. Neurosci., 35, 5023–5029.

Actively Anticipating Upcoming Tempo Changes Modulates Induced Neural Beta Power

Emily Graber 1 and Takako Fujioka 1,2

1Center for Computer Research in Music and Acoustics, Stanford University, Stanford, California, USA

2Neurosciences Institute, Stanford University, Stanford, California, USA

Investigation While playing and listening to music, musicians and listeners can not only track the beat of the music, but also anticipate upcoming tempo changes such as gradual increases or decreases in the beat rate (called ‘accelerandi’ or ‘ritardandi’). In the present study, we investigated how active temporal anticipation in listening to auditory beat stimuli is related to neural oscillations in the beta range (13-30 Hz). Typically, beta-band dynamics are characterized by decreases in power prior to and during movement (event-related desynchronization, ERD), and increases in power after movement (event-related synchronization, ERS). However, induced beta-band power modulations are also related to anticipatory temporal processing; ERS rates scale in proportion to predictable isochronous tempi, peak amplitudes are altered by temporal expectation and attention, and ERD is modulated by imagined metrical structure. We hypothesized that while listening to isochronous beat sequences, active temporal anticipation for upcoming ‘accelerandi’, ‘ritardandi’ or ‘no changes’ could alter beta power modulation patterns according to the anticipated beat timing. EEG was recorded from musicians while they listened to beat sequences at 100 bpm and anticipated gradual tempo changes according to informative cues presented before every sequence. The onset of the tempo change within a sequence was varied from trial to trial. In the ‘no change’ condition, beta power showed prototypical modulation patterns in line with previous findings (Fujioka et al., 2012). In contrast, the induced beta oscillations during directed anticipation were different prior to accelerations and decelerations in both ERD and ERS amplitudes in auditory and motor areas. In a separate experimental block, subjects identified cued tempo changes faster than uncued changes, confirming the facilitative effects of active anticipation. These results support the idea that beta synchronization and desynchronization play roles in top-down processes and that beta-band dynamics reflect active, predictive timing mechanisms.


Prediction, music, EEG, adult, 10s-100s of ms, beta band


Fujioka, T., Trainor, L. J., Large, E. W., & Ross, B. (2012). Internalized timing of isochronous sounds is represented in neuromagnetic beta oscillations. J. Neurosci., 32, 1791–1802.

Neural Entrainment Reflects Temporal Predictions Guiding Speech Comprehension

Anne Kösem 1,2

1Max Planck Institute for Psycholinguistics, Nijmegen, The Netherlands

2Radboud University, Donders Institute for Brain, Cognition, and Behaviour, Nijmegen, The Netherlands

During listening, low-frequency neural oscillations entrain to the dynamics of the speech signal, and neural entrainment has been hypothesized to be functionally relevant for speech processing. In this MEG experiment, the entrainment of oscillations in auditory cortex was manipulated to test its influence on speech perception. The manipulation was operated based on the hypothesis that neural entrainment reflects temporal predictions: the brain would internalize the rhythms of preceding signals to process the ongoing sensory input. Hence, ongoing neural oscillatory activity could be manipulated by changing the dynamics of past sensory stimulation. Using speech sentences that suddenly increased or decreased in rate, neural entrainment to past speech was expected to last after the speech rate changes, and to influence speech perception. The beginning of the sentence was either presented at a fast or a slow speech rate, while the last three words (target window) were displayed at an intermediate rate across trials. Dutch participants were asked to report the perception of the last word of the sentence, which was ambiguous with regards to its vowel duration (short vowel /ɑ/ – long vowel /aː/ contrast). Sustained neural entrainment was reported after rhythmic stimulation: brain oscillatory activity that corresponded in frequency to the preceding speech rate was observed during the target window. The sustained neural entrainment correlated with speech perceptual biases: participants who showed stronger persisting neural entrainment were more influenced by the past speech rate in their perception of the last word. These findings provide empirical support for oscillatory models of speech processing, suggesting that neural oscillations actively modulate speech comprehension.


Language, rhythms, MEG, adult, 100s of ms-secs

Visual to Auditory Entrainment Enhances Auditory Gap Detection Performance

Anna-Katharina R. Bauer 1 , Martin G. Bleichner 1,2 , Sylvain Baillet 3 and Stefan Debener 1,2

1Neuropsychology Lab, Department of Psychology, University of Oldenburg, Germany

2Cluster of Excellence Hearing4All, University of Oldenburg, Germany

3McConnell Brain Imaging Centre, Montreal Neurological Institute, McGill University, Canada

Sounds are inherently characterized by their temporally evolving and rhythmic nature. However, our sensory systems are continuously receiving many inputs that are often interrelated. Indeed, visible events often cause subsequent sounds. The synchronization of neural oscillations in the auditory cortex induced by visual rhythmic stimulation is called cross-modal phase entrainment. In two experiments using electro-encephalography (EEG) and magnetoencephalography (MEG), we investigate how we integrate cross-modal rhythmic information over time and the influence of cross-modal phase entrainment on auditory gap detection performance. Listeners were presented with auditory-only (~4.33 s) and visual-auditory stimuli (~7 s). Participants had to detect short silent gaps that were systematically distributed with respect to the phase of a 3 Hz frequency-modulated tone. In the visual-auditory condition, the visual stimulation consisted of a Gaussian pulsating circle, which preceded the auditory tone by ~2 s and was presented either in or out of phase. We found that gap detection performance increased for the visual-auditory conditions relative to the auditory-only condition and accuracies were modulated by stimulus phase. Analysis of the power spectral density revealed spectral peaks at 3 Hz and the 6 Hz harmonic at the sensor level (EEG) and in visual and auditory regions of interest (MEG). ERP analysis on the sensor level and ERF analysis in the auditory ROI revealed higher amplitudes for detected gaps. Spectral analysis revealed enhanced power in the auditory cortex during visual stimulation (MEG). We showed that cross- modal phase entrainment leads to enhanced gap detection performance and neural entrainment effects. We will further present connectivity analysis between the visual and auditory cortices to identify the mechanism underlying cross-modal phase entrainment.


Multisensory, rhythms, adult, secs-mins, MEG, EEG

Parallel Preparation of Serial Timing for Skilled Sequence Production (MEG)

K. Kornysheva 1,2 , D. Bush 2 , A. Meyer 2 , A. Sadnicka 3 , N. Burgess 2 and G. Barnes 4

1School of Psychology, Bangor University, Wales, UK

2Institute of Cognitive Neuroscience, University College London, London, UK

3Institute of Neurology, University College London, London, UK

4Functional Imaging Laboratory, University College London, London, UK

Monkey electrophysiology suggests that serial actions in skilled movements are prepared in parallel with the probability of corresponding neural pattern activations being weighted by their serial occurrence (Averbeck et al., 2002). This offers clear support for the competitive queuing account of motor sequence control. The current study aimed at testing 1) whether this effect could be reproduced in humans using non-invasive imaging (MEG) pattern classification and in how far the competitive queuing reflected the temporal structure of the planned sequences. Sixteen healthy participants were trained for two days to associate abstract visual cues with a subsequent production of the respective spatio-temporal sequence from memory following a go cue. Sequences followed a two-by-two factorial design involving a combination of two spatial and two temporal features of sequences. As in previous serial reaction time paradigms (Kornysheva et al., 2013; Kornysheva & Diedrichsen, 2014), training to produce spatio-temporal sequences from memory lead to the behavioural transfer of both spatial and temporal sequence features to new sequence combinations. The dynamic probability profiles for action specific MEG pattern activations confirmed previous electrophysiology results in monkeys suggesting parallel competitive queuing of sequential actions in the preparation period. Critically, the neuronal ordering reflected the temporal structure of sequences, independently of the specific actions employed in the sequence. Finally, the absolute probability distance between patterns related to neighboring actions during preparation predicted subsequent behavioural performance across subjects. In the high performing group the distance between adjacent action pattern probabilities during preparation was significantly higher than in the low performing group. Our results provide a fundamental insight into the neural preparation of temporally structured sequences in humans, and advocate a new neural signature of accurate preparation for skilled sequence performance.


Averbeck, B.B., Chafee, M.V., Crowe, D.A., & Georgopoulos, A.P. (2002). Parallel processing of serial movements in prefrontal cortex. PNAS, 99, 13172–13177.

Kornysheva, K., & Diedrichsen, J. (2014). Human premotor areas parse sequences into their spatial and temporal features. eLife, 3, e03043.

Kornysheva, K., Sierk, A., & Diedrichsen, J. (2013) Interaction of temporal and ordinal representations in movement sequences. J. Neurophysiol., 109, 1416–1424.

Neuronal Synchrony Mediates the Attention to Time and its Estimation

Shrikanth Kulashekhar 1,2 , Matias J. Palva 2 and Satu Palva 2

1SISSA, Trieste, Italy

2Neuroscience Center, University of Helsinki, Finland

The estimation of durations is crucial for perception, action, and cognition. Studies on timing in humans using functional magnetic resonance imaging (fMRI) have revealed a positive correlation between the estimation of multi-second temporal durations and neuronal activity in the sensory and motor areas, prefrontal and parietal cortices, basal ganglia and cerebellum (Muller & Nobre, 2014). While the brain regions involved in the estimation of durations are fairly well known, the systems-level mechanisms coordinating the distributed neuronal activity in the timing areas have remained elusive. It has been suggested that synchronized oscillations could regulate communication in neuronal networks and hence serve such coordination. However, the role of synchronized oscillations in the estimation of multi-second time intervals has remained largely unknown (Muller & Nobre, 2014). We recorded human neuronal activity using magnetoencephalography (MEG) during a temporal discrimination task adapted from Coull et al. (2004). A cue instructed the subjects to either estimate and compare the durations or in the contrast condition, the colours of dynamic visual stimuli. To address the functional significance of neuronal synchronization we identified frequency specific, large-scale synchronized networks from source reconstructed MEG data. Time estimation was associated with sustained long-range synchronization in the alpha- (10–12 Hz) and beta- (14–30 Hz) bands. We found the synchronization in-between the sensory and motor areas, prefrontal, parietal and visual cortices regions. In addition, we found that attending to the temporal aspect of the stimuli was associated with synchronization in the gamma band (40–80 Hz) between the sensory and motor areas. The results thus suggest that alpha and beta-band phase synchronization is related to the estimation of durations in the seconds’ range, while gamma-band synchronization could regulate the coordination and maintenance of temporal attention.


Duration, attention, rhythms, synchrony, MEG, adult, 100s of ms-secs


Coull, J. T., Vidal, F., Nazarian, B., & Macar, F. (2004). Functional anatomy of the attentional modulation of time estimation. Science, 303, 1506–1508.

Muller, T., & Nobre, A. C. (2014). Perceiving the passage of time: Neural possibilities. Ann. N. Y. Acad. Sci., 1326, 60–71.

Anticipation of a Familiar Event Is Associated with Widespread Predictive Activation of Sensory Cortical Areas

Dan Lloyd

Department of Philosophy and Program in Neuroscience, Trinity College, Connecticut, USA

Anticipation and expectation are familiar aspects of subjective experience, but their ubiquity in experience makes them elusive to experimental manipulation. Functional MRI datasets in the Human Connectome Project (, however, offer a way in: Seventy adult subjects were scanned at 7 Tesla while watching 60 minutes of various movie clips. One of the segments, an 88 second montage, repeats four times during the scanning session. With the fourth repetition, the montage is very familiar and depicted events can be anticipated. Anticipation then can be isolated via the following analysis, based on a comparison of the first, unanticipated, presentation of the montage, with the fourth, highly anticipated, presentation. First, we identify voxel time series that are strongly correlated between the two presentations, that is, we identify the voxels that are very strongly associated with the depicted events in both the unfamiliar (first) and familiar (fourth) presentations of the same montage. Then we search for the presence of anticipatory activation in these voxels, as follows: For each, we offset the fourth segment time series in comparison to the montage, sliding it forward in time. In other words, we ask whether the voxel-wise response to the fourth segment is similar to the first, but occurs ahead of the events in the film. In about 4% of voxels, the correlation is greater when the fourth segment voxel time series is time shifted toward the future, an effect, which is very strong at 1 second but widely present even at three seconds ahead of the expected stimulus event. These voxels are found throughout the sensory cortices, but not in other regions. Sensory response to the familiar movie stimulus is accordingly anticipatory and predictive, occurring ahead of the actual stimulus event.


Prediction, fMRI, adult, secs-mins, phenomenology, protention

Interaction of Temporal Predictions across Cues, Durations, and Behavioral Systems

Matthew S. Matell

Department of Psychology, Villanova University, USA

Identifying the mechanisms underlying temporal prediction can be facilitated by examining how different temporal expectations influence one another. We have seen averaging of temporal expectations, shared shifts in temporal expectation, and the transfer of temporal information from Pavlovian to instrumental behavior. In the first situation, rats are trained using the peak procedure that two different cues signal two different durations (e.g., tone = 10s and light = 20s). At test, presentation of the tone + light compound results in a scalar peak function at an intermediate duration (i.e., 16s), suggesting that rats are averaging their temporal memories under conditions of conflict (Swanton et al., 2009). This averaging behavior reflects the relative reliability of each cue in predicting reinforcement (Matell & Kurti, 2014), and can be blocked by administration of a serotonergic 1a antagonist at test, suggesting that averaging occurs at retrieval. In the second situation, rats are again trained that two different cues predict two different durations (i.e., tone = 8s and light = 16s). After discriminable temporal control is established, we shifted the duration associated with one of the cues (e.g., 16s light = 32s). Testing the response to the other cue demonstrates a shift in the expectation in the same direction (i.e., 8s tone = 12s; De Corte et al., submitted). This common shift results from alteration in the reinforcement time of the first cue, rather than alteration in the reinforcement rate. In the third situation, rats are trained with a Pavlovian peak procedure that a conditioned stimulus (CS) cue predicts an unconditioned food reward (US) at a specific time. After subsequent training on a variable interval procedure, presentation of the CS modifies the rate of operant responding in a temporally specific manner, thereby indicating that that temporal information is transferred from Pavlovian to operant behavioral systems.


Swanton, D. N., Gooch, C. M., & Matell, M. S. (2009). Averaging of temporal memories by rats. J. Exp. Psychol. Anim. Behav. Process, 35, 434–439.

Matell, M. S., & Kurti, A. N. (2014). Reinforcement probability modulates temporal memory selection and integration processes. Acta Psychol., 147, 80–91.

De Corte, B. J., Della Valle, R. B., & Matell, M. S. (submitted). ‘The times they are a-changin’: Temporal covariance and the common cause hypothesis.

Stimulus and Temporal Control of Behavior

Marcelo S. Caetano

Federal University of ABC (UFABC), Brazil

When trained with multiple fixed intervals (FIs), humans and other animals are able to predict the approximate time of an event and learn to discriminate different time intervals signaled by different environmental stimuli. After some training, stimulus- controlled performance emerges, which is illustrated by an earlier increase in responding for stimuli that signal shorter FIs, as compared with stimuli that signal longer FIs. However, previous work from our group has shown that, under certain conditions, stimulus control does not emerge even after hundreds of trials of training. Specifically, when signaled FIs are trained in a blocked design, performance seems to be controlled by the temporal regularity in the task, even though the stimuli still reliably signal the different FIs. Importantly, this counterintuitive result is only observed when single-trial analyses are performed, as averaged data seem to mask the effect of training design. That competition for behavior control between temporal cues and environmental stimuli can be interpreted as an instance of overshadowing, in which the temporal cues overshadow the stimuli. In this talk, I will present results from several different experiments in humans (e.g., Guilhardi et al., 2010; Labliuk et al., 2015) and rats (e.g., Caetano et al., 2012), including recent unpublished data from our lab, which describe the conditions under which stimulus- or temporally-controlled performance is observed, and show that temporal cues seem to always prevail over external cues in the competition for behavior control. The implication of those results to training designs and current models of timing will be briefly discussed.


Duration, behavior, humans, rats, secs-mins, stimulus control


Caetano, M. S., Guilhardi, P., & Church, R. M. (2012). Stimulus control in multiple temporal discriminations. Learn. Behav., 40, 520–529.

Guilhardi, P., Menez, M., Caetano, M. S., & Church, R. M. (2011). The effect of stimulus discriminability on strategies for learning multiple temporal discriminations. Behav. Processes, 84, 476–483.

Labliuk, T. P., Guilhardi, P., Cravo, A. M., Church, R. M., & Caetano, M. S. (2015). Determinants of temporal or stimulus control in humans and rats. Int. J. Comp. Psychol., 28, 1–13.

Associate Memory of Elapsed Time: Behavior, Brain Oscillations and Hippocampal Activity

Vincent van de Ven, Julia Lifanov, Oana Iosif, Sarah Kochs, Fren Smulders and Peter de Weerd

Department of Cognitive Neuroscience, Faculty of Psychology and Neuroscience, Maastricht University, The Netherlands

The extent to which time duration is represented in associate memory remains under- investigated. We designed a time paired associate task (TPAT) in which participants implicitly learnt cue-time-target associations between cue-target pairs and specific cue-target intervals ranging from 500 to 2000 msec (van de Ven et al., 2017). Importantly, participants only judged whether a cue and probe item were part of the same pair, while making no explicit judgment about time. During learning, some cue- target pairs became associated to a short interval while others became associated to a long interval. During subsequent memory testing, cue-target pairs were shown with both the short and long intervals. Participants (N = 41) showed increased accuracy of identifying matching cue-target pairs if the time interval during testing matched the implicitly learnt interval. A control experiment showed that participants had no explicit knowledge about the time associations. In subsequent neuroimaging experiments we investigated the neural correlates of TPAT memory performance. Using ultra-high field magnetic resonance imaging (UHF-MRI) study at 7 Tesla (N = 6+) we found less hippocampal activity when time intervals during test trials matched the learnt interval, compared to when they did not match. Further, in an electroencephalography (EEG) study (N = 16) we found decreased Theta oscillation power (centered at 6 Hz) at midline occipital/parietal scalp locations for the same comparison of trial types. These findings are in line with the role of hippocampus and Theta oscillations in associate memory (Buzsáki, 2006) and suggest that the same mechanisms also play a role in representing time in memory (Ranganath & Hsieh, 2016). Mismatch between presented and expected associate memory of time may change hippocampal activity and cortical Theta oscillations, possibly through a common neural source. We suggest that cue-dependent retrieval of time in associate memory could perhaps serve as a mechanism for prospective coding of expected visual spatiotemporal events.


Associate time memory, prediction, ultra-high-field fMRI, EEG, adult, 100s of ms-secs


Buzsáki, G. (2006). Rhythms of the brain. Oxford: Oxford University Press.

Ranganath, C., & Hsieh, L. T. (2016). The hippocampus: A special place for time. Ann. N. Y. Acad. Sci., 1369, 93–110.

van de Ven, V., Kochs, S., Smulders, F., & De Weerd, P. (2017). Learnt interval time facilitates associate memory retrieval. Learn. Mem.

Interval Timing in Aging Mice

Ezgi Gür 1 , Yalçın Akın Duyan 1 , Sertan Arkan 2 , Ayşe Karson 2 and Fuat Balcı 1

1Koç University, Department of Psychology, Research Center for Translational Medicine, Istanbul, Turkey

2Kocaeli University, Faculty of Medicine, Department of Physiology, Kocaeli, Turkey

The aim of this study was to investigate how interval timing behavior of mice as assessed in the dual peak procedure is altered by aging. To this end, 38 male mice (C57BL/6) at different ages (4, 10 and 18 months old) were tested in the dual peak procedure with 25s and 50s target intervals. Start, stop, middle times, and spread derived from single-trial analysis were compared between age groups for steady-state performance. Correlations between start-stop, start-spread, and middle-spread were examined to estimate the contribution of different sources of timing variability in different age groups (Gibbon & Church, 1992). Preliminary analyses showed that aged mice acquired temporally controlled response terminations later than young and adult mice. Although no differences were observed between the start times of different age groups, the stop times of old mice were significantly later than the stop times of young and adult mice at steady state. Middle times were longer than the short interval in all age groups, while it was also longer than the long interval in aged mice. Positive start-stop, negative start-spread, and positive middle-spread correlations pointed at the contribution of both clock/memory and decision processes to timing variability in all age groups. Although the obtained correlation patterns did not point at altered contributions of different sources of timing variability across age groups, the prominent differences in stop times are suggestive of the failure of behavioral inhibition in old mice. We are currently investigating the relationship between these behavioral observations and the neurobiological markers that relate to the dopaminergic and acetylcholinergic function.


Duration, behavior, aging, animal (C57BL/6 mice), secs-mins


Gibbon, J., & Church, R. M. (1992). Comparison of variance and covariance patterns in parallel and serial theories of timing. J. Exp. Anal. Behav., 57, 393–406.

Temporal Estimation of Familiar Actions in Children and Adults

Anne-Claire Rattat 1 and Valérie Tartas 2

1Laboratoire Sciences de la Cognition, Technologie, Ergonomie (SCoTE) EA 7420, Université Fédérale de Toulouse Midi-Pyrénées, INU Champollion, Albi, France

2Laboratoire Cognition, Langues, Langage, Ergonomie (CLLE) UMR 5263, Université Fédérale de Toulouse Midi-Pyrénées, Toulouse, France

Duration of event is an important aspect of temporal knowledge that children experience from a very early age. Estimating the duration of an action often requires that we reference our own knowledge about the duration of similar actions stored in long-term memory. As a consequence, the ability to encode and maintain actions’ durations in long-term memory is a keystone of time estimation. Until now, studies on memory for durations have been conducted mainly on the explicit aspect of arbitrary duration (e.g., 5-s tones) estimation and mainly with adults, and less is known about its developmental aspects (Rattat & Droit-Volet, 2007). A unique study conducted by Friedman (1990) showed that children as young as 3 years of age are able to use a scale to make relative duration judgments about daily events. The present study aims at examining memory for familiar durations by using a developmental approach, by using two different but complementary methods. In a first experiment, we examine whether or not children aged 3-, 5- and 8-years old and adults are able to categorize familiar actions according to their duration, by using a forced-choice categorization task. The results showed that with age increasing, the percentage of accurate temporal categorization increased while that of temporal categorization errors decreased. In a second experiment, 5- and 8-years old and adults have to produce the duration of familiar actions, this method being more efficient for studying memory for familiar durations in adults. The results showed that both the 8-year-olds and the adults produced longer durations for the long than for the short actions, whereas no difference was observed in the younger children. These findings are discussed in the light of event representation theory and the role of experience in temporal cognition.


Durations, children, behavior, secs-mins


Friedman, W. J. (1990). Children’s representation of the pattern of daily activities. Child Dev., 61, 1399–1412.

Rattat, A.-C., & Droit-Volet, S. (2007). Implicit long-term memory for duration in young children. Eur. J. Cogn. Psychol., 19, 271–285.

Selective Impairment of Temporal versus Spatial Predictive Coding in Schizophrenia

Valentina Ciullo 1,2 , Federica Piras 1 , Gianfranco Spalletta 1,3 and Jennifer T. Coull 4

1Neuropsychiatry Laboratory, Department of Clinical and Behavioral Neurology, IRCCS Santa Lucia Foundation, Rome, Italy

2Department of Neurosciences, Psychology, Drug Research and Child Health, University of Florence, Italy

3Menninger Department of Psychiatry and Behavioral Sciences, Baylor College of Medicine, Houston, Texas, USA

4Laboratoire des Neurosciences Cognitives, Aix-Marseille University, France & Centre National de la Recherche Scientifique, Paris, France

Timing disturbances have been proposed as an endophenotypic marker for neuropsychiatric disorders within the psychotic-affective spectrum. However, in- depth investigations of disorder-specific timing alterations are still scarce. As failures in predictive coding have been consistently observed in schizophrenia and possibly linked to the core symptom pathology, we investigated the behavioural benefits of predictive coding in the temporal versus spatial domain in 30 schizophrenic (SZ) and 30 bipolar (BD) patients, compared to matched healthy controls (HC). In an attentional orienting task, participants detected visual targets appearing at one of three locations after one of three intervals. Expectations about stimulus location (spatial condition) or onset (temporal condition) were conveyed by valid or invalid symbolic cues. Results indicated that SZ patients were able to make use of sensory information to form temporal predictions implicitly as their performance was influenced by foreperiod effects. By contrast, patients did not benefit from symbolic predictive cues (valid vs. neutral) in either the temporal or spatial dimension. Nevertheless, they were perturbed by spatially unexpected (invalid) events, though not temporally unexpected ones. Taken together, results confirm that SZ rely more upon sensory evidence to make predictions than top-down priors, and that this is particularly the case in the temporal domain. We also measured performance on temporal and colour discrimination tasks that were matched for working memory (WM) demands. Although temporal discrimination was impaired in SZ, confirming prior findings, colour discrimination was equally impaired, suggesting duration estimation impairments in SZ may be underpinned by WM dysfunction. BD patients showed no significant problems in timing compared to HC, in either the orienting or discrimination paradigms. Results therefore also challenge the theory of a common biological phenotype for SZ and BD, and provide evidence that predictive timing may serve as a model system for differentiating the specific pathophysiology of these neuropsychiatric disorders.


Prediction, duration, neuropsychology, schizophrenic patients, bipolar patients, 100s of ms-secs

Spared Synchronization to the Beat of Music in the Presence of Poor Beat Perception

Valentin Bégel 1,2 , Charles-Etienne Benoit 1,3 , Angel Correa 4,5 , Diana Cutanda 6 , Sonja A. Kotz1 7,8 and Simone Dalla Bella 1,9,10

1EuroMov, Montpellier University, Montpellier, France

2NaturalPad, SAS, Montpellier, France

3Cognition and Actions Laboratory, Institute of Neuroscience, Université Catholique de Louvain, 1200 Brussels, Belgium

4Centro de Investigacion Mente, Cerebro y Comportamiento, University of Granada, Spain

5Departamento de Psicologia Experimental, University of Granada, Spain

6Centro de Investigacion Mente, Cerebro y Comportamiento, University of Granada, Spain

7Maastricht University, Faculty of Psychology & Neuroscience, Dept. of Neuropsychology & Psychopharmacology, Maastricht, The Netherlands

8Max Planck Institute for Human Cognitive and Brain Sciences, Dept. of Neuropsychology, Leipzig, Germany

9Institut Universitaire de France (IUF), Paris, France

10International Laboratory for Brain, Music, and Sound Research (BRAMS), Montreal, Canada

Moving to the beat of music is widespread in humans. The majority, musicians and nonmusicians alike, can synchronize to the beat, for example by finger tapping or body swaying. Yet, some individuals encounter major difficulties in performing this task (poor synchronizers), a phenomenon initially ascribed to poor beat perception (i.e., beat deafness). More recently, however, it has been shown that poor synchronization can co-exist with unimpaired beat perception, thus pointing to a separation between beat perception and synchronization. Here we report the opposite dissociation (poor perception in the presence of spared synchronization) in two cases with beat deafness (L.A. and L.C). In the first experiment, participants’ rhythmic skills were assessed with the Battery for The Assessment of Auditory Sensorimotor and Timing Abilities (BAASTA), which includes both perceptual and sensorimotor tasks, and compared to that of a control group. L.A and L.C performed poorly on rhythm perception tasks, such as detecting durational shifts in a regular sequence. Yet, they could tap consistently to the beat of the same stimuli. In the second experiment, we tested whether the aforementioned perceptual deficit in L.A. and L.C. extend to an implicit timing task. In this task the regular temporal pattern of isochronous auditory stimuli aids in performing a non-temporal tasks (i.e., responding as fast as possible to a different target pitch after a sequence of standard tones). L.A. and L.C’s performance in the implicit timing task was comparable to that of controls. Altogether, the finding that synchronization to a beat can occur even in the presence of poor perception suggests that timing perception and action may dissociate at an explicit level. Spared implicit timing mechanisms might afford unimpaired synchronization to the beat in some beat-deaf participants.


Rhythm, music, neuropsychology, adult, 100s of ms-secs

Musical Rhythmic Training and Conversational Temporal Skills in Children with Deafness

Céline Hidalgo 1,2 , Simone Falk 3 , Noël Nguyen 1 and Daniele Schön 2

1Aix Marseille Univ, CNRS, LPL, Aix-en Provence, France

2Aix Marseille Univ, INSERM, INS, Inst Neurosci Syst, Marseille, France

3Laboratoire Phonétique et Phonologie, UMR 7018, CNRS/Université Sorbonne Nouvelle Paris-3, France

The smoothness of verbal exchanges during conversation depends on rhythmic coordination between interlocutors (Keller et al., 2014). The neuronal synchronization of the listener’s sensory-motor system onto the speaker’s speech rhythm gives the capacity to the listener to develop temporal predictions and anticipate speech turns. Cochlear implanted (CI) children show great variability in their linguistic development (Castellanos et al., 2015). A study has already shown the benefits of a rhythmic priming on their speech perception during a sentences repetition task (Cason et al., 2015) but no study has addressed the effect of a rhythmic training on temporal predictions at the conversational level. In three studies, we measure the temporal prediction skills of normal-hearing (NH) children and CI children during verbal interaction and the benefits of a rhythmic training versus a usual auditory training in CI children. The first study, in which children name pictures in alternation with a virtual partner, shows that the rhythmic training makes IC children sensitive to a temporal structure within the verbal exchange. Using the same type of paradigm combined with a measurement of MMN, a second study shows that NH children detect temporal deviations in the regularity of the verbal exchange. Moreover, NH children converge spontaneously towards their interlocutor on temporal parameters such as speech rate. We also demonstrate an effect of rhythmic training compared to auditory training at an electrophysiological and behavioural levels in CI children. In a third study, we assess differences between CI and NH children in anticipating turns of conversational agents during a dialog as well as the extent to which a rhythmic training can possibly enhance these temporal prediction abilities in CI children.


Prediction, music, language, electrophysiology, patients (children with deafness), 100s of ms secs


Cason, N., Hidalgo, C., Isoard, F., Roman, S., & Schön, D. (2015). Rhythmic priming enhances speech production abilities: Evidence from prelingually deaf children. Neuropsychology, 29, 102–107.

Castellanos, I., Pisoni, D. B., Kronenberger, W. G., & Beer, J. (2015). Neurocognitive function in deaf children with cochlear implants: Early development and long-term outcomes. In: The Oxford Handbook of Deaf Studies and Language: Policy, Practice, and Research (pp. 264–275).

Keller, P. E., Novembre, G., & Hove, M. J. (2014). Rhythm in joint action: psychological and neurophysiological mechanisms for real-time interpersonal coordination. Philos. Trans. R. Soc. Lond. B Biol. Sci., 369, 201