A case-by-case analysis of EPN and LPP components within a "one-picture-per-emotion-category" protocol

Abstract

Stimuli encountered in the environment are continuously evaluated according to their affective stimulus significance. Numerous event-related potential studies have shown that the early posterior negativity (EPN) and the late positive potential (LPP) are larger for high than low arousing emotional pictures. The group approach has been recently extended to the study of the individual case. Usually, many exemplars are used to represent an emotion category. Determining how many pictures are needed to reliably assess affective stimulus evaluation processes at the individual level is crucial when moving toward the goal of exploring idiosyncratic emotional stimuli. Accordingly, in the present study (N = 16), singular images displaying erotic, neutral, and mutilation content were shown 800 times while dense sensor EEG was recorded. At the group level, enhanced EPN and LPP amplitudes for high compared to low arousing stimuli emerged. At the single subject level, significantly larger amplitudes to the erotic than neutral image were observed in 15 out of 16 tests for the EPN and LPP components. Regarding the mutilation image, 15 participants showed a significant EPN effect, while the LPP effect was only found in 10 cases. Notably, emotional modulation of the EPN and LPP was stable over time. The present study contributes to the development of experimental designs tailored to the needs of the case-by-case approach. Since the process of affective stimulus evaluation is considered as a process common-to-all, the use of a singular stimulus exemplar may prove useful to investigate the idiosyncratic nature of emotion.

Keywords: EPN; ERPs; LPP; attention; case‐by‐case; emotion.

FIGURE 1

Illustration of event‐related brain potential (ERP) waveforms and difference scalp maps (high–low arousal) of the emotional modulation of the early posterior negativity (EPN) and late positive potential (LPP). Grand mean ERP waveforms were calculated using the respective clusters displayed in the sensor layouts (with the nose pointing up). The upper panels showing a back view of the model head illustrate the EPN findings which appear as a relative negative shift for high compared to low arousing images. The lower panels, showing a top view of the model head (with the nose pointing up), illustrate the LPP effect, appearing as a positive shift for highly arousing images. The scalp maps represent the average difference across a time window from 192–252 ms (EPN) and 380–480 ms (LPP).

FIGURE 2

Emotional modulation of the early posterior negativity (EPN) and late positive potential (LPP) appeared stable across stimulus repetitions. Each colored dot represents the single trial EPN/LPP activity, averaged across participants. Colored lines represent a linear regression trend.

FIGURE 3

Illustration of the single subject findings for the early posterior negativity (EPN). (a) The upper panel shows difference scalp maps (high‐low arousal) of the emotional modulation of the EPN for each individual case. A back view of the model head is displayed. Maps show the individual time window of the maximal EPN difference as selected by the automatic software routine. Please note the common scale (±1.5 μV), which leads to truncation of effects in individuals showing large effects. (b) The lower panel illustrates the case‐by‐case statistics. The bootstrap distribution is shown for each case in a boxplot. The bottom and top edges of the box indicate the 25th and 75th percentiles while the whiskers indicate the 5th and 95th percentiles. The dots illustrate the de facto measured EPN difference (high–low arousal). Dots outside the range indicated by the whiskers represent significant effects (p < .05) with size and color indicating different p‐levels.

FIGURE 4

Illustration of the single subject findings for the late positive potential (LPP). (a) The upper panel shows difference scalp maps (high‐low arousal) of the emotional modulation of the LPP for each individual case. A top view of the model head is displayed (with the noise pointing up). (b) The lower panel illustrates the case‐by‐case statistics. For more details, please refer to Figure 3.

FIGURE 5

Proportion (%) of significant cases as a function of p‐ criterion and number of trials used for the calculation of the bootstrap sampling distribution. Blue and red colors indicate the comparison of erotica‐neutral and mutilation‐neutral, respectively.