Solving the “Cocktail Party Problem:” Fluctuating Background Noise Effects on Signal Recognition in Cope’s Grey Tree Frog

Additional contributors: Alejandro Vélez; Sandra Tekmen; Mark Bee (faculty mentor)If you have ever had trouble understanding what a person was saying to you in a noisy group setting, then you have experienced a condition known as “the cocktail-party problem.” The cocktail party problem is a condition in which an individual has trouble hearing and understanding an isolated target signal in a noisy group setting. When natural increases and decreases in the levels of background noise occur, a phenomenon known as “modulation masking release” may allow the listener to decipher the target signal during the times when background noise is lowest. Frogs breed in noisy groups known as choruses. Within these choruses, females must isolate the most attractive male call possible from the cacophony around them—thus, they are also subject to the cocktail party problem. Using simulated mating calls and modulated chorus noise, my colleagues and I attempted to study how grey tree frogs may use masking release to resolve the cocktail party problem within mating groups. I hypothesized that exposure to low-frequency modulated chorus-like noise would allow frogs to experience modulation masking release, resulting in an overall improved phonotaxis (sound response) compared to unmodulated (constant) levels of background noise. Additionally, I expected the effect of masking release to be more prominent when the mating call was longest. However, experimental results revealed that the presence of modulated chorus noise confers no significant benefit to females when they are isolating and choosing male calls with respect to flat, unmodulated noise. Future successful studies such as this one could ultimately benefit the hearing-impaired by helping audiologists to create better hearing technology.This research was supported by the Undergraduate Research Opportunities Program (UROP)