Repetition suppression for pitch using fMRI

Perception of communication signals in noisy environments is a critical function of the auditory system. Our ability to perceive a communication signal in a noisy environment is aided by certain properties of the background noise: hearing thresholds in co-modulated noise are lower than in unmodulated noise. This effect is called “co-modulation masking release”. It is hypothesized that the fluctuating nature of the co-modulated background gives the auditory system access to “glimpses” of the stimulus, therefore enabling lowered detection thresholds. While this effect, along with other forms of masking and release from masking, has been studied extensively in humans, to date, there have not been any studies using non-human primates. Here, we describe a novel tone-in-noise detection task. Monkeys were trained to report the presence of a tone embedded in either co-modulated or unmodulated noise. On a trial-by-trial basis, the audibility of the tone was varied systematically by changing the intensity of the tone, relative to a background noise with a fixed intensity. When the monkey heard the tone, he released a lever to obtain a juice reward. Behavioral performance was quantified with a psychophysical curve and a d’ analysis. We report how performance varied as a function of tone frequency, tone intensity, and noise type (co-modulated or unmodulated). These results provide a novel behavioral task that can be used to study the link between hearing in noisy environments and its neural underpinnings. Primary sensory cortex predicts the utility of specific sensory information in a behavioral learning set by enhancing the cortical representation of the critical signal. K. M. BIESZCZAD*, & N. M. WEINBERGER Cntr Neuro. of Learn. & Memory and Dept. of Neuro. & Behav., Univ. California Irvine, Irvine, CA. The representation of acoustic frequency in the tonotopic map of primary auditory cortex (A1) is a convenient model for “item-specific” information storage in the cerebral cortex. For example, tonotopy in A1 encodes the behavioral relevance of a signal frequency by enhancing its cortical representation: the importance of the acoustic-frequency detail of an experience can be represented in the number of A1 cells that become closely tuned to that frequency (Rutkowski & Weinberger, 2005) which may increase its strength memory (Bieszczad & Weinberger, 2010). Here we investigated how A1 might represent the importance of more than one frequency that have become signals in a learning set. We examined the dynamics of receptive field plasticity as rats implanted with an array of recording electrodes learned to bar-press for rewards to one tone-signal (5 kHz) and in a second phase, to discriminate between this and another spectrally-distant signal-frequency (5 vs. 12 kHz). Multiple recordings from A1 were made prior to daily training sessions to characterize frequency response areas (FRAs) at each site. FRAs were compared to daily performance to investigate neural predictive correlates of behavioral frequencydiscriminations. We report the development of A1 receptive field plasticity in a behavioral challenge, “partial reversal discrimination”, when an original single CS+ (5 kHz), was made the unrewarded CSwith a new CS+ (12 kHz) introduced for discrimination. A frequencydiscrimination learning set was established by subsequent consecutive reversals in which the signal-frequency that was rewarded (CS+) or unrewarded (CS-) was reversed. The findings reveal that current reward valence and prior experience combine to reflect highorder instantiation of frequency-specific A1 plasticity that serves the function of memory for useful auditory signals.