The effect of perceptual load on serial dependence

The experiment was conducted in a quiet, light-shielded room. Participants were seated 57 cm from the screen and viewed the stimuli with their head stabilized by a chin-and-forehead rest to control viewing distance and minimize head movements.Behavioral Data Analysis: To examine serial dependence in the orientation task, the current study conducted analyses from both the "previous stimulus" and "previous response" perspectives.

First, the relative orientation between the current and previous trials was calculated separately for each approach. For the previous-stimulus analysis, the relative orientation (x_stimulus) was defined as the orientation of the preceding stimulus minus the orientation of the current stimulus, with positive values indicating that the previous stimulus was more clockwise than the current stimulus. For the previous-response analysis, the relative orientation (x_response) was defined as the participant's reported orientation on the previous trial minus the orientation of the current stimulus, with positive values indicating that the previous response was more clockwise than the current stimulus.

Next, the adjustment error for the current trial was computed as the difference between the reported orientation and the actual presented orientation, with positive values denoting a clockwise adjustment error. Finally, the adjustment error was modeled as a function of the relative orientation separately for the previous-stimulus and previous-response analyses. The first derivative of a Gaussian function (DoG) was fitted to the adjustment errors for each participant under high and low perceptual load conditions, yielding an amplitude parameter.

Larger amplitude values indicate a stronger serial dependence effect. The DoG function is expressed as follows:y=h+(x+b)awce-(w(x+b))²where x is the relative orientation between consecutive trials (x_stimulus or x_response); a is the curve amplitude, representing the strength of serial dependence; w is the width; h is the height; b is the intercept; and c is a constant.To compare the strength of serial dependence between high and low perceptual load conditions, paired-sample t-tests were conducted on the amplitude values derived from the previous-stimulus and previous-response analyses separately.

In addition, to verify the effectiveness of the perceptual load manipulation, paired-sample t-tests were performed on the mean reaction time and accuracy of the letter-search task, as well as the mean reaction time of the orientation-judgment task, between the high and low perceptual load conditions.