Time Window


Sounds differ in the duration over which information is conveyed. For instance, phonemes are short in duration while syllables are much longer. So the optimal duration of time window required for analysis is different for different acoustic features. But how does the brain organise the processing of sound segments that require use of time-windows of different duration? Is this anatomical organization of time-window processing as seen in humans consistent with other primates?

Visual summary

​Here is a visual summary of this project. 

I explored whether monkeys are a good model of human brain mechanisms underlying processing of time windows. I conducted fMRI based BOLD imaging in awake behaving macaques using synthetic stimuli to show that their anatomical organization of processing of time windows is similar to humans. However, monkeys exhibit a decreased sensitivity to longer time windows as compared to humans. 

This difference in sensitivity between humans and monkeys is surprising given their phylogenetic proximity. This difference in sensitivity for long time windows between species might be due to the specialization of the human brain for processing of speech which requires greater sensitivity to longer time windows. My study highlights the brain mechanisms that might be unique to humans, possibly an outcome of divergent evolution alongside the development of speech.


Here is a poster summarising the project


Source code

The MATLAB based source code for generating Spectral Flux stimulus is shared below

Pradeep D., Kumar, S., & Griffiths, T. (2020, September 28). MATLAB scripts for generating Spectral Flux stimuli. 


Relevant publication

The peer reviewed publication is shared below