Multisensory integration in the ventral intraparietal area of the macaque monkey

Abstract

The goal of this study was to characterize multisensory interaction patterns in cortical ventral intraparietal area (VIP). We recorded single-unit activity in two alert monkeys during the presentation of visual (drifting gratings) and tactile (low-pressure air puffs) stimuli. One stimulus was always positioned inside the receptive field of the neuron. The other stimulus was defined so as to manipulate the spatial and temporal disparity between the two stimuli. More than 70% of VIP cells showed a significant modulation of their response by bimodal stimulations. These cells included both bimodal cells, i.e., cells responsive to both tested modalities, and seemingly unimodal cells, i.e., cells responding to only one of the two tested modalities. This latter observation suggests that postsynaptic latent mechanisms are involved in multisensory integration. In both cell categories, neuronal responses are either enhanced or depressed and reflect nonlinear sub-, super-, or additive mechanisms. The occurrence of these observations is maximum when stimuli are in temporal synchrony and spatially congruent. Interestingly, introducing spatial or temporal disparities between stimuli does not affect the sign or the magnitude of interactions but rather their occurrence. Multisensory stimulation also affects the neuronal response latencies of bimodal stimuli. For a given neuron, these are on average intermediate between the two unimodal response latencies, again suggesting latent postsynaptic mechanisms. In summary, we show that the majority of VIP neurons perform multisensory integration, following general rules (e.g., spatial congruency and temporal synchrony) that are closely similar to those described in other cortical and subcortical regions.

Figure 1.

Single-cell examples of multisensory integration in bimodal neurons. Rasters and response peristimulus time histograms to tactile (left), visual (middle), and bimodal (right) stimulation. Bars on raster represent spikes, and rows indicate trials. Neuronal responses are aligned to stimulus onset (gray line). Peristimulus time histograms are the summed activity across all trials in a given stimulus condition (bin width of 15 ms). Tick lines at the bottom of the peristimulus time histograms represent stimulation duration. A–C, Single-cell examples of enhancement responses. A, The neuronal activity is significantly higher in the bimodal condition compared with the visual condition (t test, p = 0.05), and multisensory integration takes the form of a sub-additive enhancement (amplification index, +17%; additive index, −5%). B, Mean firing rate histogram of the same cell as in A of the visual (V), tactile (T), bimodal (VT), and arithmetical sum of visual and tactile (V+T) responses (t test, *p ≤ 0.05; **p ≤ 0.01). The dashed line represents spontaneous firing rate. C, Example of a super-additive enhancement effect in a bimodal neuron (t test, p = 0.05; amplification index, +30%; additive index, +21%). D–F, Single-cell examples of depression responses. D, Visual–tactile neuron whose tactile response is significantly depressed (t test, p = 0.01) by the concurrent presentation of a visual stimulus (amplification index, −20%; additive index, −33%). E, This integrative response falls between the tactile and the visual response. F, Example of a VIP neuron, for which the bimodal response is inferior to the minimal unimodal response (amplification index, −60%; additive index, −70%).

Figure 2.

Single-cell examples of multisensory integration in unimodal neurons. A, B, Single-cell examples of enhancement responses. A, Unimodal tactile neuron (t test; tactile response, p = 0.001; visual response, p = 0.41) whose tactile response is significantly enhanced when combined with ineffective visual information (t test, p = 0.05; amplification index, +36%; additive index, +36%; the indices are identical because it is a unimodal cell: V+T will thus be equal to Uni_max, and both indices are equivalent). B, Mean firing rate histogram of the same neuron. C, D, Single-cell examples of depression responses. C, Unimodal visual neuron (t test; visual, p = 0.001; tactile, p = 0.47) whose visual response is significantly depressed when combined with an ineffective tactile information (t test, p = 0.05; amplification index, −45%; additive index, −45%; here also the two indices are equal, for the same reasons as described above). D, Mean firing rate histogram of the same neuron. Same conventions as in Figure 1. V, Visual; T, tactile; VT, bimodal; V+T, arithmetical sum of visual and tactile.

Figure 3.

Distribution of the amplification and the additive indices in the entire population of recorded cells (n = 150). Gray circles illustrate unimodal neurons (n = 63), and black triangles indicate bimodal neurons (n = 87). On the top and on the right of the scatter plot are shown the specific distribution plots for each index. The amplification index is derived from the work of Meredith and Stein (1986a,b), whereas the additive index has been adapted from the work of Populin and Yin (2002). The amplification index reflects either an increase or a decrease of the discharge to a bimodal stimulation with respect to the discharge evoked by a visual or tactile stimulus alone (see Materials and Methods). The additive index compares the bimodal response to the predicted linear sum of the response to each single sensory modality. A negative [respectively (resp.) positive] index characterizes a sub-additive (resp. super-additive) response. Moreover, an amplification (resp. additive) index of 33% corresponds to a bimodal response that is twice the maximal unimodal response (resp. the arithmetical sum), an index of 50% corresponds to a bimodal response that is three times the maximal unimodal response (resp. the arithmetical sum), and an index of 66% corresponds to a bimodal response that is five times the maximal unimodal response (resp. the arithmetical sum).

Figure 4.

Multisensory integration effects on neuronal latency. A, Single-cell example. Bimodal response (blue spike density function) is significantly shorter than visual response (green) (one-way ANOVA, p = 0.01) and significantly longer than tactile response (yellow) (one-way ANOVA, p = 0.01). B, Distribution of tactile (yellow bars), visual (green), and visuo-tactile (blue) latencies in the population of bimodal integrative and non-integrative cells (n = 81). Arrows point to the mean tactile, visual, and bimodal latencies.

Figure 5.

Effects of stimulus time onset asynchrony on multisensory integration in VIP. A, Histogram plot of a single-cell example of the mean firing rate as a function of SOA (error bars indicate mean ± SE). Single-modality responses of the cell are shown on the left part of the histogram (visual, dark gray bar; tactile, light gray bar). On the right part, bimodal responses to different SOA conditions are represented by black bars. Dashed line corresponds to the spontaneous activity. Multisensory integration is significant only when bimodal stimulations are synchronous (t test, *p = 0.05; amplification index, +34%; additive index, +6%). B, Population distribution of maximal integrative effects in function of stimulus onset asynchrony (n = 37). Same color conventions as in A. C, Population distribution of temporal window sizes in which multisensory integration can be found across the recorded population (n = 37). Same color conventions as in A. T, Tactile; V, visual.

Figure 6.

Effect of spatial congruence on multisensory integration in VIP. A, Single-cell example, spatially congruent configuration (left) and incongruent configuration (right). At the top, the tactile response of the neuron and a drawing of the tactile receptive field location (shaded area) on a monkey face are shown. The tactile stimulus is always presented inside the tactile receptive field (Tin) in both configurations. The visual stimulus is either presented inside (Vin) or outside (Vout) the visual receptive field (middle panel, left and right, respectively). Schematic representation of the location of the visual stimulation is shown. At the bottom, the neuronal responses to congruent bimodal stimulations (Vin Tin, left) and incongruent bimodal stimulations (Vout Tin, right). B, Population distribution of multisensory integration effects as a function of spatial congruence/incongruence (n = 65).

Figure 7.

Inverse effectiveness principle in VIP. The amplification index is plotted as a function of the dominant unimodal response for the subset of integrative neurons with enhancive multisensory responses (n = 43). Enhanced bimodal integration increases when the maximal unimodal response decreases.