Temporal Electroencephalography Traits Dissociating Tactile Information and Cross-Modal Congruence Effects

Abstract

To explore whether temporal electroencephalography (EEG) traits can dissociate the physical properties of touching objects and the congruence effects of cross-modal stimuli, we applied a machine learning approach to two major temporal domain EEG traits, event-related potential (ERP) and somatosensory evoked potential (SEP), for each anatomical brain region. During a task in which participants had to identify one of two material surfaces as a tactile stimulus, a photo image that matched ('congruent') or mismatched ('incongruent') the material they were touching was given as a visual stimulus. Electrical stimulation was applied to the median nerve of the right wrist to evoke SEP while the participants touched the material. The classification accuracies using ERP extracted in reference to the tactile/visual stimulus onsets were significantly higher than chance levels in several regions in both congruent and incongruent conditions, whereas SEP extracted in reference to the electrical stimulus onsets resulted in no significant classification accuracies. Further analysis based on current source signals estimated using EEG revealed brain regions showing significant accuracy across conditions, suggesting that tactile-based object recognition information is encoded in the temporal domain EEG trait and broader brain regions, including the premotor, parietal, and somatosensory areas.

Keywords: EEG; classification analysis; event-related potentials; object recognition; somatosensory evoked potentials.

Figure 1

Experimental design. (A) Characteristics of two materials with different degrees of surface smoothness (i.e., bumpy and slippery) used in the experiment. For tactile stimuli, one of the physical materials was presented to the right index finger, and for visual stimuli, one of the photo images of the materials was presented on a PC display. (B) A table summarizing eight combinations of attention, tactile, and visual stimuli in the two experimental conditions, congruent and incongruent. (C) Time flow for a session. One trial consisted of four periods that lasted 6–8 s in total depending on the lengths of the rest period. In the second period, one of the two target stimuli, tactile or visual, was shown on the screen for 1 s so that the participants would pay attention to the instructed stimulus in the next period. In the third period of 3 s, a photo image of the two materials was presented on the screen, and one of the physical materials was presented to the corresponding participant’s index finger. Electrical stimulation was also applied to the right wrist median nerve throughout the period. In the fourth period, the participants verbally reported which material they were touching within 1 s. One session included 40 trials consisting of 5 trials for eight individual combinations in random order. One participant engaged in three sessions in total. (D) An example trial of stimulus presentation. This example shows an incongruent condition in which the bumpy surface is presented as a visual stimulus and the slippery surface is presented as a tactile stimulus during tactile attention. When visual and tactile stimuli were presented, electrical stimulation was also presented to evoke SEP. During the stimulus presentation time, visual and tactile stimulation were presented continuously but electrical stimulation was presented discretely.

Figure 2

EEG settings. (A) The EEG setup. (B) Electrode placement based on the international 10–20 method.

Figure 3

Classification accuracies between conditions using current-source-based ERP and SEP are displayed in each brain region. The classification was performed in the following areas: M1, S1, S2, PL, SMA, PM, ACC, PCC, SMG, PC, IPL, IPS, SPL, POS, and Occ. The three-second signal after the initiation of material presentation was divided into three periods (0–1 s, 1–2 s, and 2–3 s after the material was presented), and classification was performed for each time period. The classification accuracy of the shuffled labels (baseline) was used for comparison with congruent and incongruent conditions.

Figure 4

Brain regions showing significant accuracy differences compared to the chance level calculated using shuffled labels (congruent condition (A) and incongruent condition (B)) and significant differences between the congruent and incongruent conditions (C). The S1, PL, SMA, PM, ACC, PCC, SMG, PC, IPL, IPS, SPL, POS, and Occ areas showed a significant difference between the congruent and incongruent conditions. (D) Location of each brain region.