Mamiko Niwa

Perceptual Decisions between Multiple Directions of Visual Motion

Previous studies and models of perceptual decision making have largely focused on binary choices. However, we often have to choose from multiple alternatives. To study the neural mechanisms underlying multialternative decision making, we have asked human subjects to make perceptual decisions between multiple possible directions of visual motion. Using a multicomponent version of the random-dot stimulus, we were able to control experimentally how much sensory evidence we wanted to provide for each of the possible alternatives. We demonstrate that this task provides a rich quantitative dataset for multialternative decision making, spanning a wide range of accuracy levels and mean response times. We further present a computational model that can explain the structure of our behavioral dataset. It is based on the idea of a race between multiple integrators to a decision threshold. Each of these integrators accumulates net sensory evidence for a particular choice, provided by linear combinations of the activities of decision-relevant pools of sensory neurons.

Differences between Primary Auditory Cortex and Auditory Belt Related to Encoding and Choice for AM Sounds

We recorded from middle–lateral (ML) and primary (A1) auditory cortex while macaques discriminated amplitude-modulated (AM) noise from unmodulated noise. Compared with A1, ML had a higher proportion of neurons that encoded increasing AM depth by decreasing their firing rates (“decreasing” neurons), particularly with responses that were not synchronized to the modulation. Choice probability (CP) analysis revealed that A1 and ML activity were different during the first half of the test stimulus. In A1, significant CP began before the test stimulus, remained relatively constant (or increased slightly) during the stimulus, and increased greatly within 200 ms of lever release. Neurons in ML behaved similarly, except that significant CP disappeared during the first half of the stimulus and reappeared during the second half and prerelease periods. CP differences between A1 and ML depend on neural response type. In ML (but not A1), when activity was lower during the first half of the stimulus in nonsynchronized, decreasing neurons, the monkey was more likely to report AM. Neurons that both increased firing rate with increasing modulation depth (“increasing” neurons) and synchronized their responses to AM had similar choice-related activity dynamics in ML and A1. These results suggest that, when ascending the auditory system, there is a transformation in coding AM from primarily synchronized increasing responses in A1 to nonsynchronized and dual (increasing/decreasing) coding in ML. This sensory transformation is accompanied by changes in the timing of activity related to choice, suggesting functional differences between A1 and ML related to attention and/or behavior.

Task Engagement Selectively Modulates Neural Correlations in Primary Auditory Cortex

Noise correlations (rnoise) between neurons can affect a neural populations discrimination capacity, even without changes in mean firing rates of neurons. rnoise, the degree to which the response variability of a pair of neurons is correlated, has been shown to change with attention with most reports showing a reduction in rnoise. However, the effect of reducing rnoise on sensory discrimination depends on many factors, including the tuning similarity, or tuning correlation (rtuning), between the pair. Theoretically, reducing rnoise should enhance sensory discrimination when the pair exhibits similar tuning, but should impair discrimination when tuning is dissimilar. We recorded from pairs of neurons in primary auditory cortex (A1) under two conditions: while rhesus macaque monkeys (Macaca mulatta) actively performed a threshold amplitude modulation (AM) detection task and while they sat passively awake. We report that, for pairs with similar AM tuning, average rnoise in A1 decreases when the animal performs the AM detection task compared with when sitting passively. For pairs with dissimilar tuning, the average rnoise did not significantly change between conditions. This suggests that attention-related modulation can target selective subcircuits to decorrelate noise. These results demonstrate that engagement in an auditory task enhances population coding in primary auditory cortex by selectively reducing deleterious rnoise and leaving beneficial rnoise intact.

Amplitude modulation detection as a function of modulation frequency and stimulus duration: Comparisons between macaques and humans

Previous observations show that humans outperform non-human primates on some temporally-based auditory discrimination tasks, suggesting there are species differences in the proficiency of auditory temporal processing among primates. To further resolve these differences we compared the abilities of rhesus macaques and humans to detect sine-amplitude modulation (AM) of a broad-band noise carrier as a function of both AM frequency (2.5 Hz-2 kHz) and signal duration (50-800 ms), under similar testing conditions. Using a go/no-go AM detection task, we found that macaques were less sensitive than humans at the lower frequencies and shorter durations tested but were as, or slightly more, sensitive at higher frequencies and longer durations. Humans had broader AM tuning functions, with lower frequency regions of peak sensitivity (10-60 Hz) than macaques (30-120 Hz). These results support the notion that there are species differences in temporal processing among primates, and underscore the importance of stimulus duration when making cross-species comparisons for temporally-based tasks.

Hierarchical effects of attention on amplitude modulation encoding in auditory cortex

How attention influences single neuron responses in the auditory system remains unresolved. We found that when monkeys actively discriminated temporally amplitude modulated (AM) from unmodulated sounds, primary auditory (A1) and middle lateral belt (ML) cortical neurons better discriminated those sounds than when the monkeys were passively listening. This was true for both rate and temporal codes. Differences in AM responses and effects of attentional modulation on those responses suggest: (1) attention improves neurons’ ability to temporally follow modulation (2) non-synchronized responses play an important role in AM discrimination (3) ML attention-related increases in activity are stronger and longer-lasting for more difficult stimuli consistent with stimulus specific attention, whereas the results in A1 are more consistent with multiplicative nonlinearity, and (4) A1 and ML code AM differently; ML uses both increases and decreases in firing rate to encode modulation, while A1 primarily uses activity i...