The midstream order deficit

Abstract

Warren, Obusek, Farmer, and Warren (1969) reported a striking inability of listeners to perceive the relative order of a sequence of four sounds presented repeatedly, despite their fairly accurate performance when the sequence was presented only once. Their participants first heard the four sounds separately and learned a name for each. Then they were told that the four sounds would be presented in a particular order repeatedly and that afterwards they would be asked to report the correct relative order. A correct relative order report could begin with any one of the four sounds. For example, if buzz-beep-hiss-boop was presented over and over, hiss-boop-buzz-beep would be one correct response. The sounds were played for 200 msec each, with no interstimulus interval, and the sequence was repeated continuously without pauses between repetitions. The sequence ended when the participants were ready to respond. However, they felt that they could not apprehend the order, no matter how long they listened, and, in fact, their performance was not significantly different from chance. In contrast, another group of participants to whom the sequence was presented just once, instead of multiple times, performed significantly better than chance.In the three present experiments, we first demonstrate a similar effect with visual stimuli and then test several candidate explanations for the effect. Our results suggest that poor performance with cycling sequences is not simply due to a greater difficulty in perceiving the individual items; rather, cycling causes a problem that is specific to the apprehension of order. In particular, we found that the accuracy of order judgments was correlated with the degree to which an identifiable item was perceptually distinct from the preceding items. We suggest that this was due to a difficulty in rapidly initiating order encoding without having a salient perceptual event. EXPERIMENT 1 Visual Letters With Single Versus Multiple PresentationThis experiment documents conditions in which accuracy at judging relative order is lower for repeated presentations than for a single presentation of a sequence. We roughly equated forward and backward masking in the This research was supported by NIH Grant MH45245 to N.K., NSF Grant SBR-9511633 to A.T., an NIMH NRSA graduate fellowship to A.O.H., and NIH Grant EY092582 to Patrick Cavanagh. We are indebted to Rik Henson for sharing his knowledge of theories of shortterm order memory. We thank Etienne Benson, Jared Bush, and Damian Stanley for their help in conducting the experiments, and Charles Stromeyer, Patrick Cavanagh, and two anonymous reviewers for their useful comments on earlier manuscripts. Portions of this research were reported at the Annual Meeting of the Association for Research in Vision and Ophthalmology, 1996, Ft. Lauderdale, FL, The relative order of an auditory sequence can be more difficult to apprehend when it is presented repeatedly without pause (i.e., cycling) than when it is presented only once (Warren, Obusek, Farmer, & Warren, ...

Attentional awakening: gradual modulation of temporal attention in rapid serial visual presentation

“…The results showed that correct identiWcations for earlier target positions (the Wrst and second Bins) were lower, and were consistently higher after the third Bin. This Wnding is consistent with the hypothesis based on attentional preparation, or the initiating act suggested by Holcombe et al (2001). In the beginning of the RSVP sequence, the observers would begin attentional preparation to eYciently process a target that might appear anytime.…”

“…In all the experiments, there was a transient deWcit in identifying a target presented at the very beginning of the RSVP sequence. This deWcit in the identiWcation of the target could be the cost of the process of orienting temporal attention, the attentional preparation, which is consistent with the implication of Holcombe et al (2001). The attentional preparation process in the Fig.…”

“…Given this assumption, even if there is no information about when relevant events will occur, people could have some attentional preparation for unpredictable relevant events that might facilitate task performance. This idea is consistent with an implication by Holcombe, Kanwisher, and Treisman (2001) using the rapid serial visual presentation (RSVP) task, in which temporal information is implicitly available for performing the task. The RSVP task requires observers to identify a target item deWned by a certain feature among a rapid stream of distractor items (e.g., a white target letter among lightblue distractor letters).…”

“…To test this issue, we systematically manipulated the target's temporal position within the RSVP sequence. If attentional preparation for a target that might appear anytime occurs as an initiating act, as Holcombe et al (2001) suggested, then observers should not correctly report the target inserted in the beginning of the RSVP sequence, but would correctly report the target later in the sequence.…”

Independent, synchronous access to color and motion features

“…No, because unlike in our ring condition, with only a half-cycle stimulus snippet presented, a color-motion asynchrony is still found, as previously reported by Linares and López-Moliner (2006) and as found in a pilot experiment we conducted using the stimulus parameters of this experiment (data not shown). Our ring condition also differs from presentation of only a half-cycle in that a half-cycle presentation introduces temporal order cues including visual persistence and onset cues that do not occur in a longer stream (Beaudot, 2002;Dakin & Bex, 2002;Holcombe, Kanwisher, & Treisman, 2001) even when the longer stream is exogenously cued (Holcombe, Kanwisher, & Treisman, unpublished observation).…”