Asymmetric interference in concurrent time-to-contact estimation: Cousin or twin of the psychological refractory period effect?

Abstract

In a reaction time (RT) task requiring fast responses to two stimuli presented close in time, human observers show a delayed RT to the second stimulus. This phenomenon has been attributed to a psychological refractory period (PRP). A similar asymmetric interference is found when performing multiple concurrent visual time-to-contact (TTC) estimations for moving objects, despite important differences between the tasks. In the present study, we studied the properties of the asymmetric interference found in the TTC task and compared them to the classical PRP effect. In Experiment 1, we varied the time interval between the two objects' arrival times to determine the dependence of the PRP-like effect on the asynchrony between the two TTCs. In Experiment 2, we investigated whether the physical or the perceived arrival order determined the asymmetric interference. Our results confirmed the existence of asymmetric interference in the multiple TTC estimation task, but also indicated important differences from the traditional PRP effect observed in the RT paradigm. The origins of these differences are discussed, as well as the practical implications.

Asymmetrical time-to-contact error with two moving objects persists across different vertical separations

“…While temporal proximity is undoubtedly a key factor in the asymmetric pattern of error found when making two concurrent TTC estimations in the PM task, here we examined if there was also an influence of vertical separation between the two objects. Consistent with Baurès et al (2010Baurès et al ( , 2011Baurès et al ( , 2017, we found that temporal estimation was significantly more accurate for the lead object than the trail object. Extending upon previous work, we also found that the overestimation in CE for the trail object compared to the lead object was similar across a range of TTCs.…”

“…Although the reference and distractor objects had identical visual features and an equal probability of moving at 5 or 7.5 deg/s in the upper or lower vertical location, TTC of the reference object was fixed at 1900 ms, whereas TTC of the distractor varied by ±250 ms or ±500 ms. Therefore, the pattern of CE described above and reported in previous work (Baurès et al, 2010;2017) could be specific to TTC of the reference object, which was constant across all trial types. To examine this issue, we repeated the same analysis on CE of the distractor object.…”

“…We used two approaches for quantifying error in TTC estimation. Similar to Baurès et al (2010Baurès et al ( , 2011Baurès et al ( , 2017, we first calculated constant error (CE) for each of the two objects relative to their respective arrival time (i.e., 1900ms for the reference, 1400, 1650, 2150 or 2400 ms for the distractor).…”

“…This situation requires the participant to make two concurrent TTC estimations and results in an asymmetrical pattern of error. Participants exhibit the expected level of accuracy for the object that arrives first (i.e., lead object) but significantly overestimate TTC of the second object when it trails (the lead object) by a short temporal delay (Baurès, DeLucia, & Olson, 2017).…”

“…In the current study, therefore, we conducted two experiments that examined the influence of vertical separation between two moving objects on accuracy of TTC estimation. In Experiment 1, we replicated the object features used in previous work (Baurès et al, 2010;2017), whereas in Experiment 2 we modified the shape of one object in order to facilitate disambiguation. Importantly, the evolving horizontal separation between the two objects was dependent on their respective velocity and actual TTC, and thus would not independently account for any differences as a function of vertical separation.…”