Human posterior parietal cortex responds to visual stimuli as early as peristriate occipital cortex

Abstract

Much of what is known about the timing of visual processing in the brain is inferred from intracranial studies in monkeys, with human data limited to mainly noninvasive methods with lower spatial resolution. Here, we estimated visual onset latencies from electrocorticographic (ECoG) recordings in a patient who was implanted with 112 subdural electrodes, distributed across the posterior cortex of the right hemisphere, for presurgical evaluation of intractable epilepsy. Functional MRI prior to surgery was used to determine boundaries of visual areas. The patient was presented with images of objects from several categories. Event-related potentials (ERPs) were calculated across all categories excluding targets, and statistically reliable onset latencies were determined, using a bootstrapping procedure over the single trial baseline activity in individual electrodes. The distribution of onset latencies broadly reflected the known hierarchy of visual areas, with the earliest cortical responses in primary visual cortex, and higher areas showing later responses. A clear exception to this pattern was a robust, statistically reliable and spatially localized, very early response, on the bank of the posterior intraparietal sulcus (IPS). The response in the IPS started nearly simultaneously with responses detected in peristriate visual areas, around 60 ms poststimulus onset. Our results support the notion of early visual processing in the posterior parietal lobe, not respecting traditional hierarchies, and give direct evidence for onset times of visual responses across the human cortex.

Keywords: ECoG; early visual processing; electrocorticography; onset latency estimation.

Fig. 1. Illustration of the onset latency estimation method.

A Calculation of onset latency estimate (OLE) for a specific electrode via bootstrapping over the baseline trials, construction of empirical distributions of maximal and minimal t-values, and using the 0.01 percentile values as thresholds for the response t-value signal (see methods). B. Estimation of temporal error of the OLE. Examples of specific electrodes having small or large temporal errors (left and right, respectively). Shaded blue area is 99% confidence interval around the mean. Red lines are the translation of the constant t-value thresholds back to the voltage domain, multiplying by the instantaneous standard error (see methods for detailed explanation). Electrode numbers are specified in Fig. 2.

Fig. 2. Significant visual onset latencies in all electrodes.

Four views of the same brain are presented. The color of the circles denotes the timing of the estimated onset (see color bar). The size of the circles denotes the temporal error estimate (see methods): larger circles indicate more reliable estimates (smaller temporal errors; see legend). White dots indicate electrodes that were labeled as epileptic or with other electric artifacts. However, we looked for onset latencies of the bad electrodes as well, and if a significant onset was found, the colored circle was placed under the white dot (e.g. electrode 78). Black dots indicate electrodes for which no significant onset was detected. Numbers are indicated here for specific characteristic electrodes, and electrodes that are referred to in the text. Note that some electrodes appear in more than one view, however their onsets and/or numbers might appear in some of the views, e.g., the onsets of inferior electrodes are not indicated in the posterior view. Reference: common average (CAR). Electrodes that are explicitly mentioned in the text are labeled on the posterior and inferior views. For a dynamic depiction of activation over time see supporting information video S1.

Fig. 3. Posterior view of early onset latency estimates.

The same as in figure 2, but with magnified temporal-scale, see color bar. Note that onset latencies estimated later than 100 ms are colored with the darkest blue on the scale. Labels according to retinotopic mapping as in figure 5 (see methods). Size of the circles is due to temporal error estimate, as in figure 2.

Fig. 4. Event-related single trials in early responding electrodes.

Electrode 68 (left) and electrode 34 (right). Electrodes are labeled from retinotopic mapping (see methods), as in figures 3 and 4. Note the remarkably consistent onsets in individual trials.

Fig. 5. Waveforms and Onset Latency Estimates (OLEs).

Comparing electrodes having retinotopic labels, using Common Average Reference (CAR), Current Source Density (CSD) or broadband gamma power (Gamma). To the left of each row, the number and retinotopic label of each electrode is specified. For each method, the scale is specified in the first row. Insets for broadband gamma power signals, electrodes 34 and 67, show a magnified y-scale in order to better view the responses. Missing electrodes in CSD were excluded from the analysis because they reside on the edge of the occipital strip and therefore are not appropriate for CSD estimation. Missing electrodes in the broadband gamma power case did not show a significant OLE (see Table 1)