Tackling the functional role of multisensory interactions in behavioural benefits: A combined EEG and computational modelling approach

Supervisors: Justin Ales, Thomas Otto

Project Description:
The availability of multiple senses is highly beneficial to control behaviour. Different senses not only increase the spectrum of perceivable signals but also provide redundant signals that, by combination, enable better estimates of external events and/or faster achievements of subjective goals. However, we do not yet fully understand the brain functions that guide the combination of multisensory signals. Strikingly, a number of recent studies have demonstrated the involvement of early sensory cortices in multisensory processes, which in many cases included even primary sensory cortices that have traditionally been considered to maintain unisensory processes only. Still, the exact functional role of these early multisensory interactions and their contribution to behavioural benefits remain puzzling (Kayser, 2010).

The proposed PhD project will tackle multisensory interactions in humans by an interdisciplinary approach. The project will utilize a new computational model of multisensory integration to interpret recordings from state-of-the-art steady-state visual evoked potential (SSVEP) recordings (Norcia et al., 2015). The SSVEP provides for the unique ability to frequency-tag input presented to the different senses. By tagging the input to each sense it is possible to track the flow of information from unisensory input to multisensory interactions. These tags retain unambiguous signatures of their sensory origin even after being combined by multisensory processes. Second, we will analyse the neurophysiological data informed by a new computational modelling approach, which allows for precise predictions about the processing interactions that have contributed to the benefits observed in behaviour and the underlying perceptual decisions (Otto & Mamassian, 2012). 

The combination of advanced physiological recording with cutting-edge computational modelling will enable an unprecedented understanding of how the brain combines information across the senses.

Research Training

The PhD project is suitable for students in Neurosciences (and related disciplines including Psychology, Biology, Physics, and Computer Sciences). The successful candidate will have research interests in human sensory processes and perceptual decision making. Prior experience in EEG and/or computational modelling is a plus but is not a requirement, as research training in both disciplines will be provided.

The PhD candidate involved in this project will develop a thorough understanding of multisensory integration, state-of-the-art electrophysiological techniques and a solid understanding of experimental design practices. The techniques learned during this project (including programming in MATLAB and computational modelling) easily lend themselves to both further scientific studies and to more applied approaches. These techniques also provide a strong grounding in statistical analysis, advanced mathematics, and computer programming, all of which are highly relevant skills in today’s world: both in the industry and in academia.


Kayser (2010). The Multisensory Nature of Unisensory Cortices: A Puzzle Continued. Neuron, 67(2), 178-180.

Norcia AM, Appelbaum LG, Ales JM, Cottereau BR, Rossion B (2015) The steady-state visual evoked potential in vision research: A review. J Vis 15:4.

Otto & Mamassian (2012). Noise and correlations in parallel perceptual decision making. Current Biology, 22(15), 1391-1396.