# Neural basis of vocal signal recognition during natural communication

> **NIH NIH R01** · UNIVERSITY OF CALIFORNIA, SAN DIEGO · 2023 · $543,086

## Abstract

Project Summary
One of the most frequently discussed challenges facing the auditory system is parsing the voice of a single
speaker in a noisy acoustic environments comprising multiple other speakers, commonly referred to as the
Cocktail Party Problem (CPP). Despite the significance of the CPP to illustrate the complex nuances of
audition that unfold in real-world situations, remarkably little remains known about the supporting neural
mechanisms at different levels of the auditory circuit. This gap in our knowledge has emerged, at least in part,
due to a lack of a behavioral paradigm in animal models that both recreated the natural complexities and
dynamics of natural cocktail parties, while at the same time affording experimental control to manipulate the
acoustic scene. Here we seek to bridge this gap by leveraging a novel, multi-speaker behavioral paradigm in a
series of complementary experiments to investigate the perceptual and neural mechanisms that are employed
in the primate brain to resolve the CPP for natural communication in freely-moving common marmosets.
 Aim 1 establishes an innovative behavioral paradigm that recreates the natural dynamics of real-world
cocktail parties by creating an acoustic scene comprising multiple computer-generated virtual monkeys (VM)
whose individual identity and spatial position can be independently and systematically manipulated. Each
marmoset placed in these cocktail parties naturally learn the identity of the interactive VM based on
idiosyncrasies of their voice and vocal behavior and selectively engage them in conversations, while ignoring
the distractor VMs engaged in conversations with each other.
 Experiments in Aim 2 involve performing simultaneous neurophysiological recordings of prefrontal
cortex and hippocampus in marmoset monkeys as they communicate in a cocktail party described in Aim 1.
Specifically, experiments are designed to test the respective and complementary roles of these neural
substrates in the auditory representing the identity and spatial position of individuals in the CPE. Because of
the respective role of prefrontal cortex in attention and hippocampus for allocentric representations of space,
these substrates are hypothesized to be critical to resolving the CPP in primates.
 Aim 3 aims to build on the complementary conceptual and experimental innovations to more directly
interrogate the perceptual and neural mechanisms that support communicating in a cocktail party. Specifically
we will a cutting-edge closed-loop feedback paradigm that uses machine learning algorithms to optimize the
conversational dynamics between the interactive VM and live marmoset in a multi-speaker cocktail party. This
model will selectively manipulate targeted features within the ongoing natural conversations to directly
interrogate the system and determine the functional role of behavioral parameters and related neural
processes.

## Key facts

- **NIH application ID:** 10659667
- **Project number:** 2R01DC012087-11
- **Recipient organization:** UNIVERSITY OF CALIFORNIA, SAN DIEGO
- **Principal Investigator:** CORY T MILLER
- **Activity code:** R01 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2023
- **Award amount:** $543,086
- **Award type:** 2
- **Project period:** 2012-07-01 → 2028-06-30

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/10659667

## Citation

> US National Institutes of Health, RePORTER application 10659667, Neural basis of vocal signal recognition during natural communication (2R01DC012087-11). Retrieved via AI Analytics 2026-05-25 from https://api.ai-analytics.org/grant/nih/10659667. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*