Hallucination of moving objects revealed by a dynamic noise background

Abstract

We show that on a dynamic noise background, the perceived disappearance location of a moving object is shifted in the direction of motion. This “twinkle goes” illusion has little dependence on the luminance- or chromaticity-based confusability of the object with the background, or on the amount of background motion energy in the same direction as the object motion. This suggests that the illusion is enabled by the dynamic noise masking the offset transients that otherwise accompany an object’s disappearance. While these results are consistent with an anticipatory process that pre-activates positions ahead of the object’s current position, additional findings suggest an alternative account: a continuation of attentional tracking after the object disappears. First, the shift was greatly reduced when attention was divided between two moving objects. Second, the illusion was associated with a prolonging of the perceived duration of the object, by an amount that matched the extent of extrapolation inferred from the effect of speed on the size of the illusion (~50 ms). While the anticipatory extrapolation theory does not predict this, the continuation of attentional tracking theory does. Specifically, we propose that in the absence of offset transients, attentional tracking keeps moving for several tens of milliseconds after the target disappearance, and this causes one to hallucinate a moving object at the position of attention.

The Importance of Attention in the Twinkle Goes Effect

“…The results of Experiment 2a challenged the predictions of tracking continuation theory, and to an extent, predictive extrapolation theory. Strikingly, these results also contrast with previous empirical evidence from Nakayama and Holcombe (2020) that the magnitude of the twinkle goes effect is reduced when attending to four objects, compared to two, in an alignment judgement task. It is feasible that the results of Experiment 2a were due to random variation, or due to the specific characteristics of the sample.…”

“…These results also differed from Nakayama and Holcombe's (2020) original finding that tracking four objects was associated with a smaller twinkle goes effect than when tracking two, in an alignment judgement task. However, Nakayama and Holcombe analysed their data at the group-level, as opposed to the individual level.…”

“…The absence of twinkle goes effect in two out of five participants in Experiment 1 was not predicted by either predictive extrapolation or tracking continuation theory, nor was it suggested by the prior empirical research of Nakayama and Holcombe (2020). From the perspective of tracking continuation theory, it is possible that this task was too attentionally demanding for a twinkle goes effect to occur in all participants, even in the one target condition.…”

“…With the object tracking literature divided, a new illusion from Nakayama and Holcombe (2020) presents an exciting opportunity to test the implications of predictive and non-predictive theories of object tracking. Originally demonstrated through an alignment judgement task, the twinkle goes effect (Figure 2A; Supplementary Movie 1) describes a shift in the perceived location of a moving object in the direction of motion, when dynamic noise is present at the time of disappearance.…”

“…Conversely, predictive extrapolation theory does not implicate attention in the forward displacement caused by extrapolation, presenting a testable difference between these two accounts. In an initial attempt to consider the role of attention, Nakayama and Holcombe (2020) provided evidence that the twinkle goes effect is smaller when individuals perform an alignment task while attending to four objects compared to two (Figure 2B). However, this experiment used rotational motion, despite the greater attentional constraints for rotational compared to linear motion (Thornton & Gilden, 2001;.…”