In lieu of an abstract, here is a brief excerpt of the content:

19 What Determines Simultaneity and Order Perception? Piotr Jaśkowski All physical objects in the universe are embedded in three-dimensional space, and all physical events are immersed in time. Therefore, to properly act, human beings evolved abilities to orient in space and to put events in temporal order. It seems quite easy to say that an event occurred earlier than another one. However, this task becomes more difficult when the events occur very close in time to the other. It is rather obvious that in judging the order of two events the human mind has to rely on some brain events evoked by the external stimuli. It is, therefore, at least theoretically possible that, for some physiological reasons, two physical events are erroneously perceived in a reverse order. In the last three decades much effort has been put into uncovering the physiological factors underpinning the perception of the subjective temporal order. In this chapter, we review the main findings concerning this topic. 19.1 Measuring Sensory Latency with TOJ/SJ Two methods have most frequently been adopted to study perception of order and simultaneity : the temporal-order judgment (TOJ) task and the simultaneity judgment (SJ) task. Let us consider two stimuli, Sx and Sy, stimulating two separate sensory channels, x and y, and presented at tx and ty, respectively. The temporal interval between the onsets of the stimuli, tx - ty , is called stimulus onset asynchrony (SOA) and usually varies from trial to trial. In the SJ task, participants are required to judge whether or not the two stimuli were presented simultaneously. The frequency of a specific response (“simultaneous” or “nonsimultaneous”) varies systematically with SOA. One can easily imagine that if two physically identical stimuli are presented simultaneously at some distance, they should also be perceived as simultaneous. Therefore, for SOA = 0 ms, the participants will report “simultaneous” most frequently. If SOA deviates more and more from zero, this response will be less and less frequent. Thus, the psychometric function for SJ should be a bell-shaped curve with maximum for SOA = 0. The SOA at which the psychometric function reaches its maximum is called the point of subjective simultaneity (PSS). For two identical stimuli, we expect PSS = 0 ms. In the TOJ task, participants are asked to specify which of two stimuli, separated by an SOA, occurred first. If Sx (Sy) is presented much earlier than Sy (Sx) (i.e., SOA >> 0), then it is 380 Piotr Jaśkowski obvious that participants would report “Sx first” (“Sy first”) in all trials. The psychometric function (percentage of “Sx first” responses as a function of SOA) is typically fitted with a cumulative Gaussian distribution. Once SOA is close to zero, observers perceive the stimuli as simultaneous and are unable to report their order. Therefore, the SOA at which participants report equally often “Sx first” as “Sy first” should correspond to the PSS. It is worth noting that a PSS = 0 is expected only under particular conditions, namely when the two stimuli are identical and stimulate two identical sensory channels, a situation difficult to achieve. Indeed, PSS has proven to depend on many factors, which we will talk about later in this chapter. 19.2 Simultaneity versus Order Judgment So far, we accepted the assumption that both TOJ and SJ provide identical results, in the sense that PSSs obtained are equal under the same stimulation conditions. This assumption is based on seemingly obvious reasoning, which is accepted in the majority of models: if observers perceive two stimuli as nonsimultaneous, they should be able to tell the order of the stimuli. In fact, this is not as trivial as it seems. There are some empirical findings suggesting rather the reverse. For example, Jaśkowski (1991b) highlighted some peculiarities of the psychometric functions when the so-called ternary response paradigm was used (Stelmach & Herdman, 1991; Ulrich, 1987). In this paradigm, participants responded with three possible responses: “Sx first,” “Sy first,” “Sx and Sy simultaneous.” The psychometric function modeling the probability of “Sx first” (“Sy first”) as a function of the SOA was found to be nonmonotonic, having a local minimum. Jaśkowski (1991b) showed that this extraordinary behavior could be accounted for by assuming that there are two mechanisms working in parallel. One of them is responsible for telling apart “Sx first” from “Sy first,” while the second assesses whether the two stimuli were simultaneous or not. This nonmonotonicity occurred, according to this model, because in some trials people perceived the stimuli as nonsimultaneous...

Share