Effects of trial history on cross-modal non-spatial inhibition of return

Previous laboratory studies have shown that an individual’s response in the current trial can be influenced by a previous trial, and this has been described as an effect of trial history. Existing studies have shown that there is a trial history effect with visual spatial inhibition of return (IOR), and some studies have shown that changes in stimulus modalities also affect reaction times (RTs). The present study used the “prime-neutral cue-target” paradigm to examine the trial history effect in cross-modal, non-spatial IOR and attempted to decrease the trial history effect. In two experiments, we mainly manipulated the cue-target modalities in the current trial (auditory-visual vs. visual-auditory modalities), cue validity in the current trial (cued vs. uncued) and cue validity in the previous trial (cued vs. uncued). Thirty participants were recruited in Experiment 1. The visual prime cue was a red or blue disk with a radius of 2° visual angle, and the auditory prime cue was a verbal sound in Chinese at 75 dB (\hong\ or \lan\). The visual neutral cue was a green disk with a radius of 2° visual angle, and the auditory neutral cue was a verbal sound in Chinese at 75 dB (\lv\); The visual target was a red or blue disk with a radius of 2° visual angle, and the auditory target was a verbal sound in Chinese at 75 dB (\hong\ and \lan\). During the experiment, each trial began with a 400 ms fixation cross in the centre of the monitor, and a 300 ms visual or auditory prime cue was followed by a 200 ms fixation cross. After the 300 ms visual or auditory neutral cue, another fixation cross was presented for 300 ms, and then a 300 ms auditory or visual target was presented. The participants were asked to discriminate the identity of the target(i.e., either a colour disk or vocalization of \hong\or \lan\) within 1500 ms. Following a 1500 ms intertrial interval (ITI) with a blank screen, the next trial was initiated. Twenty-nine participants were recruited in Experiment 2, the ITI was 4500 ms, and the other parameters were identical to those in Experiment 1. Regarding the RTs results, Experiment 1 showed that the RTs for cued targets in the current trial were larger than RTs for uncued targets, which was a colour-based non-spatial IOR. The IOR effect size in the current trial showed an interaction between the cue validity in the previous trial and the cue-target modality in the current trial. The IOR effect size on the current trial after a valid cue trial was larger than the IOR effect size with an invalid cue in the previous trial when the current trial was a visual cue and auditory target; however, there was no difference in the IOR effect size when the cue was auditory, and the target was visual in the current trial. Furthermore, the analysis of the target modality across trials revealed that the valid cue, but not the invalid cue, in the previous trial, could induce a larger IOR effect size in the current trial with visual cues. A longer ITI (4500 ms) was used in Experiment 2 compared to Experiment 1, and the results showed that there was a difference in the IOR effect size in the current trial between the visual cues and auditory cues in the current trial. The IOR effect size in the current trial was not influenced by the validity of the previous trial or whether the current trial had auditory cues or visual cues. These results suggested an interaction between trials on cross-modal non-spatial IOR, but the effect was related to the cue-target modality. There was not only the cue validity effect across trials but also the target modality switch effect between trials. Increasing the time interval between trials can reduce the effect of the previous trial on the IOR effect size in the current trial.