Pinpointing the neural signatures of single-exposure visual recognition memory

Memories of the images that we have seen are thought to be reflected in the reduction of neural responses in high-level visual areas such as inferotemporal (IT) cortex, a phenomenon known as repetition suppression (RS). We challenged this hypothesis with a task that required rhesus monkeys to report whether images were novel or repeated while ignoring variations in contrast, a stimulus attribute that is also known to modulate the overall IT response. The monkeys’ behavior was largely contrast invariant, contrary to the predictions of an RS-inspired decoder, which could not distinguish responses to images that are repeated from those that are of lower contrast. However, the monkeys’ behavioral patterns were well predicted by a linearly decodable variant in which the total spike count was corrected for contrast modulation. These results suggest that the IT neural activity pattern that best aligns with single-exposure visual recognition memory behavior is not RS but rather sensory referenced suppression: reductions in IT population response magnitude, corrected for sensory modulation.

Distributed Sampling-based Bayesian Inference in Coupled Neural Circuits

The brain performs probabilistic inference to interpret the external world, but the underlying neuronal mechanisms remain not well understood. The stimulus structure of natural scenes exists in a high-dimensional feature space, and how the brain represents and infers the joint posterior distribution in this rich, combinatorial space is a challenging problem. There is added difficulty when considering the neuronal mechanics of this representation, since many of these features are computed in parallel by distributed neural circuits. Here, we present a novel solution to this problem. We study continuous attractor neural networks (CANNs), each representing and inferring a stimulus attribute, where attractor coupling supports sampling-based inference on the multivariate posterior of the high-dimensional stimulus features. Using perturbative analysis, we show that the dynamics of coupled CANNs realizes Langevin sampling on the stimulus feature manifold embedded in neural population responses. In our framework, feedforward inputs convey the likelihood, reciprocal connections encode the stimulus correlational priors, and the internal Poisson variability of the neurons generate the correct random walks for sampling. Our model achieves high-dimensional joint probability representation and Bayesian inference in a distributed manner, where each attractor network infers the marginal posterior of the corresponding stimulus feature. The stimulus feature can be read out simply with a linear decoder based only on local activities of each network. Simulation experiments confirm our theoretical analysis. The study provides insight into the fundamental neural mechanisms for realizing efficient high-dimensional probabilistic inference.

Pinpointing the neural signatures of single-exposure visual familiarity

Memories of the images that we have seen are thought to be reflected in the reduction of neural responses in high-level visual areas such as inferotemporal (IT) cortex, a phenomenon known as repetition suppression (RS). We challenged this hypothesis with a task that required rhesus monkeys to report image familiarity while ignoring variations in contrast, a stimulus attribute that is also known to modulate the overall IT response. The monkeys’ behavior was largely contrast-invariant, contrary to the predictions of the RS encoding scheme, which could not distinguish response familiarity from changes in contrast. However, the monkeys’ behavioral patterns were well predicted by a linearly decodable variant in which the total spike count is corrected for contrast modulation. These results suggest that the IT neural activity pattern that best aligns with single-exposure visual familiarity behavior is not RS but rather “sensory referenced suppression (SRS)”: reductions in IT population response magnitude, corrected for sensory modulation.

Neural Processing of Amplitude-Modulated Sounds

Joris, P. X., C. E. Schreiner, and A. Rees. Neural Processing of Amplitude-Modulated Sounds. Physiol Rev 84: 541–577, 2004; 10.1152/physrev.00029.2003.—Amplitude modulation (AM) is a temporal feature of most natural acoustic signals. A long psychophysical tradition has shown that AM is important in a variety of perceptual tasks, over a range of time scales. Technical possibilities in stimulus synthesis have reinvigorated this field and brought the modulation dimension back into focus. We address the question whether specialized neural mechanisms exist to extract AM information, and thus whether consideration of the modulation domain is essential in understanding the neural architecture of the auditory system. The available evidence suggests that this is the case. Peripheral neural structures not only transmit envelope information in the form of neural activity synchronized to the modulation waveform but are often tuned so that they only respond over a limited range of modulation frequencies. Ascendingthe auditory neuraxis, AM tuning persists but increasingly takes the form of tuning in average firing rate, rather than synchronization, to modulation frequency. There is a decrease in the highest modulation frequencies that influence the neural response, either in average rate or synchronization, as one records at higher and higher levels along the neuraxis. In parallel, there is an increasing tolerance of modulation tuning for other stimulus parameters such as sound pressure level, modulation depth, and type of carrier. At several anatomical levels, consideration of modulation response properties assists the prediction of neural responses to complex natural stimuli. Finally, some evidence exists for a topographic ordering of neurons according to modulation tuning. The picture that emerges is that temporal modulations are a critical stimulus attribute that assists us in the detection, discrimination, identification, parsing, and localization of acoustic sources and that this wide-ranging role is reflected in dedicated physiological properties at different anatomical levels.

Comparison at a Distance

The visual system is known to contain hard-wired mechanisms that compare the values of a given stimulus attribute at adjacent positions in the visual field; but how are comparisons performed when the stimuli are not adjacent? We ask empirically how well a human observer can compare two stimuli that are separated in the visual field. For the stimulus attributes of spatial frequency, contrast, and orientation, we have measured discrimination thresholds as a function of the spatial separation of the discriminanda. The three attributes were studied in separate experiments, but in all cases the target stimuli were briefly presented Gabor patches. The Gabor patches lay on an imaginary circle, which was centred on the fixation point and had a radius of 5 deg of visual angle. Our psychophysical procedures were designed to ensure that the subject actively compared the two stimuli on each presentation, rather than referring just one stimulus to a stored template or criterion. For the cases of spatial frequency and contrast, there was no systematic effect of spatial separation up to 10 deg. We conclude that the subject's judgment does not depend on discontinuity detectors in the early visual system but on more central codes that represent the two stimuli individually. In the case of orientation discrimination, two naïve subjects performed as in the cases of spatial frequency and contrast; but two highly trained subjects showed a systematic increase of threshold with spatial separation, suggesting that they were exploiting a distal mechanism designed to detect the parallelism or non-parallelism of contours.

Effectiveness of Redundant Color Coding in Multidimensional Identification

This study investigated the effectiveness of redundant color coding in multidimensional identification. Statistical analysis showed that redundant color in multidimensional identification did not necessarily improve performance and response speed might even deteriorate if subjects were not informed of the use of redundant color. Merely informing subjects of the use of redundant color might not benefit identification speed either; subjects had to actually use color in responding to facilitate response speed. Further, redundant color might be more appropriate to associate with the less-salient or the less-familiar stimulus attribute. Implications of the results for the design of multidimensional display and for human information processing were discussed.

Nature and precision of temporal coding in visual cortex: a metric-space analysis

1. We recorded single-unit and multi-unit activity in response to transient presentation of texture and grating patterns at 25 sites within the parafoveal representation of V1, V2, and V3 of two awake monkeys trained to perform a fixation task. In grating experiments, stimuli varied in orientation, spatial frequency, or both. In texture experiments, stimuli varied in contrast, check size, texture type, or pairs of these attributes. 2. To examine the nature and precision of temporal coding, we compared individual responses elicited by each set of stimuli in terms of two families of metrics. One family of metrics, D(spike), was sensitive to the absolute spike time (following stimulus onset). The second family of metrics, D(interval), was sensitive to the pattern of interspike intervals. In each family, the metrics depend on a parameter q, which expresses the precision of temporal coding. For q = 0, both metrics collapse into the "spike count" metric D(Count), which is sensitive to the number of impulses but insensitive to their position in time. 3. Each of these metrics, with values of q ranging from 0 to 512/s, was used to calculate the distance between all pairs of spike trains within each dataset. The extent of stimulus-specific clustering manifest in these pairwise distances was quantified by an information measure. Chance clustering was estimated by applying the same procedure to synthetic data sets in which responses were assigned randomly to the input stimuli. 4. Of the 352 data sets, 170 showed evidence of tuning via the spike count (q = 0) metric, 294 showed evidence of tuning via the spike time metric, 272 showed evidence of tuning via the spike interval metric to the stimulus attribute (contrast, check size, orientation, spatial frequency, or texture type) under study. Across the entire dataset, the information not attributable to chance clustering averaged 0.042 bits for the spike count metric, 0.171 bits for the optimal spike time metric, and 0.107 bits for the optimal spike interval metric. 5. The reciprocal of the optimal cost q serves as a measure of the temporal precision of temporal coding. In V1 and V2, with both metrics, temporal precision was highest for contrast (ca. 10-30 ms) and lowest for texture type (ca. 100 ms). This systematic dependence of q on stimulus attribute provides a possible mechanism for the simultaneous representation of multiple stimulus attributes in one spike train. 6. Our findings are inconsistent with Poisson models of spike trains. Synthetic data sets in which firing rate was governed by a time-dependent Poisson process matched to the observed poststimulus time histogram (PSTH) overestimated clustering induced by D(count) and, for low values of q, D(spike)[q] and D(intervals)[q]. Synthetic data sets constructed from a modified Poisson process, which preserved not only the PSTH but also spike count statistics accounted for the clustering induced by D(count) but underestimated the clustering induced by D(spike)[q] and D(interval)[q].