A Comparative Study of Voice Perception in Primates

Project Information

COVOPRIM

Grant agreement ID: 788240

DOI

10.3030/788240

Project closed

EC signature date 19 July 2018

Start date 1 January 2019

End date 31 December 2025

Funded under

• EXCELLENT SCIENCE - European Research Council (ERC)

Total cost

€ 2 900 000,00

EU contribution

€ 2 900 000,00

Coordinated by

UNIVERSITE D'AIX MARSEILLE
France

Periodic Reporting for period 5 - COVOPRIM (A Comparative Study of Voice Perception in Primates)

Reporting period: 2024-11-01 to 2025-12-31

Summary of the context and overall objectives of the project

COVOPRIM proposes to reconstruct the recent evolutionary history of one important component of speech and language: voice perception. The perceptual and neural mechanisms of voice perception are being compared between humans, macaques and marmosets –two highly vocal and extensively studied species of Old-World and New-World monkeys –to quantify cross-species differences and infer mechanisms potentially inherited from a common ancestor.
This research is important for society because it is expected to significantly advance our understanding of how humans uniquely evolved speech and language, and precise the neural mechanisms—shared with other primates as well as uniquely-human—involved in processing the wealth of information in voice.
The overall objectives are:
(1) to use large-scale behavioural testing based on ad-lib access of monkeys to automated systems (following the highly successful model developed locally with baboons) in several behavioural experiments that will establish psychometric response functions for robust cross-species comparison (WP1);
(2) to use comparable functional MRI scanning protocols in the humans, macaques and marmosets to compare the organization of what I hypothesize constitutes a “voice patch system” similar to the face patch system of visual cortex and broadly conserved in primates (WP2);
(3) to use fMRI-guided microstimulation during fMRI scanning to establish the effective connectivity within the voice patch system, as well as during task performance to compare the behavioural effects of voice patch microstimulation to those of transcranial magnetic stimulation of the voice areas in humans.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

Between January 2019 and June 2020 I have:
− secured Ethics approval for the Macaque and Marmoset parts of the project (all WPs);
− recruited 10 staff for carrying out WP1 and WP2;
− performed a first set of behavioural experiments in baboons; results confirm an ability by baboons to learn a voice/non-voice discrimination task and establish psychometric curves;
− setup the marmoset colony and laboratory, and initiated work with automated testing chairs (WP1);
− advanced on the setup of the macaque automated tests and programming (WP1); the macaque enclosure is now fitted with custom designed macaque testing boxes;
− advanced on the setup of the macaque and marmoset fMRI experiments (WP2); specifically I purchased, received and tested one of the only existing prototypes of 32-channel head coil for macaque MRI.

Between June 2020 and December 2021 I have:
- finalized two additional behavioural experiments in baboons; results refine the initial results and are being written as part of a PhD thesis (Fatimzahra Ennaji) (WP1);
- setup the automated behavioural experiments in marmosets; hundreds of trials are collected every days and the animals and are currently being trained at same/different auditory discrimination task (WP1);
- finalized the setup of the automated testing systems for macaques; experiments have started on a small group of macaques (2 purchased by the project, and others belonging to the primatology station) (WP1);
- finalized a first fMRI study comparing humans and macaques (WP2); manuscript published in Current Biology;
- advanced on the setup of fMRI experiments for marmosets (WP2); specifically I have purchased the most advanced multi-array head coil (16 channels) for marmosets for 3T scanner and tests have been performed satisfactorily; in parallel a 4-channel head coil has been developed with a local company for 7T scanner.
- implanted two macaques with multi-electrode arrays in fMRI-localized voice patches, and performed extensive recordings between January and July with stimuli for 4 different experiments (WP3).

Since January 2022 I have:
- Supervised to graduation 4 PhD students working on the project (Dr Fatimazhra Ennaji, Dr Manon Obliger-Debouche, Dr Charly Lamothe, Dr Mélina Cordeau)
- Completed planned data acquisition for automated behavioural tests in macaques and marmosets (WP1). Analyses are underway. Completed acquisition of marmoset vocalizations and prepared a Nature Scientific Data paper, under review.
- Completed planned acquisition of fMRI data in marmoset (article written and submitted). Completed planned acquisition of fMRI data in macaques completes, analyses under way. Stared planning acquisition of fMRI data in humans (WP2).
- Completed planned acquisition of electrophysiological data in 3 macaques. Analyses ongoing. 1 article published (PNAS), 2 more being submitted.

Since Novembre 2024:
WP1: Finalized analysis of macaque behavioural data, and completed acquisition and analysis of human behavioural data in comparable conditions. An article is being drated.
WP2: Acquisition of human and macaque fMRI data completed. Analyses and write up being finalized. Acquisition of marmoset fMRI data in collaboration with U Western Ontario; 1 article published in Current Biology.
WP3: Analyses for several macaque electrophysiology datasets under way. 1 article published in Science Advances. 1 large review article written and accepted in Trends in Cognitive Sciences.

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

This first phase of the project has seen good progress towards establishing the marmoset groups and develop automated testing systems for macaques and marmosets.
The first result beyond the state of the art has been obtained in an already established group of baboons thanks to external collaboration; it confirms the ability of baboons to learn a voice/nonvoice discrimination task, and establishes psychometric curves for future cross-species comparisons. This suggests the planned experiment in macaques and marmosets will yield interesting results as well. We now are in a od position to perform the planned behavioural, fMRI and electrophysiological experiments of the project, for a comprehensive test of my hypothesis of a primate ‘voice patch system’ shared to a large extent between primates.

Since June 2020, we have advanced well on WP1 for all 3 species: experiments are finalized in baboons, well under way in marmosets, and starting in macaques. We are in a good position to perform the planned experiments over the next years of the project.
In WP2 we have finalized a first comparative fMRI study and revealed important new results: the direct comparison of fMRI data obtained in humans and monkeys using the same scanner and auditory stimuli shows that the two species possess areas of the anterior temporal lobe that not only prefer conspecific vocalizations but also categorize them apart from other sounds in a species-specific, but functionally homologous manner. This suggests that the higher-level auditory cortex of macaques and humans is more similarly organized than currently known. The results have been published in Nov 2021 (Bodin et al, 2021).

IN WP3 we have for the first time implanted chronic multi-electrode arrays in the voice patches of two monkeys in cortical locations previously identified by fMRI. The large amount of data we have collected in these animals, currently under analysis, suggests important new results will be obtained.
Since January 2020 we have acquired data for 5 more experiments, and started analysing this treasure trove of data. The first article form WP3 has been published in PNAS in summer 2024, a second in Science Advances in 2025. Two additional papers are being prepared.