The Theory And Experimental Research On Psychological Refractory Period Effect

When human are required to respond to two stimuli presented in rapid succession, The stimulus onset asynchrony(SOA) between the two stimuli is varied, Often found as the SOA is decreased, Task 1(T1) and Task 2(T2) overlapped highly, the reaction times to the second task (RT2) increased greatly, usually by several hundred milliseconds. This form of dual- task interference, known as the psychological refractory period (PRP) effect, has been found with a wide range of tasks, including very simple ones. PRP effect depicted when two tasks were presented with rapid succession, two tasks were processed simultaneously or serially. PRP effect means human beings are plainly subject to severe limitations in their ability to perform more than one task at the same time. Because the phenomena appear to reflect a severe limitation on human parallel task performance, it has been the subject of intensive empirical and theoretical interest.There are two general types of models have been proposed to account for these delays effect. One is Pashler's response—selection bottleneck model (RSB model). the other is Tombu and Jolicoeur's central capacity sharing model (CCS model). RSB model propose that some processing needed to perform each task requires access to one or more processors that can only act on one input at a time (Pashler, 1994a). If both tasks require one of these processors simultaneously, then only one can get access to it. While this processor is busy with one task, processing for the other task must be suspended until the processor is free. We refer to processors that can only operate on one task at a time as bottleneck processors or bottleneck stages. CCS model begins with the assumption that there are stages of processing that are not capacity limited and others that are. As in most bottleneck models, the present model assumes that capacity—limited stages occur centrally. Like previous capacity sharing models, the present model assumes that the capacity limitations of the central stages are not all or none. When two tasks overlapped, the limited capacity will be allocated between two tasks. When one task accepted more capacity, the other will have no sufficient capacity, which result two tasks delayed in short SOA. CCS model considered RSB model is a special case of the central capacity sharing model in which all capacity is allocated to Task 1 for response selection and than the other when both tasks require central processing.Although RSB model and CCS model depicted PRP effect from different aspect, both models predicted different processing. At the same time both models predicts all of the hallmark effects of the psychological refractory period paradigm on RT2. Such as—1 slope of the PRP effect at short stimulus onset asynchronies (SOA), underadditivity of precentral Task 2 manipulations, additivity of central or postcentral Task 2 manipulations with SOA, and carry forward to Task 2 of Task 1 precentral or central manipulations at short SOA. But the two models made different prediction on RT1. RSB model predicts that RT1 is constant across SOA, while the CCS model predicts that Task 1 response times increase with decreasing SOA, the slope of RT1 depends on the difficulty of T2.So far, the research of PRP effect is only a limited range tasks have been examined. It is too early to make precise generalizations concerning the cognitive operations for which this bottleneck mechanism is required. In the present research will use mental rotation paradigm material, through behavior data and ERP data to examine the PRP effect, because the mental rotation material is a useful tool to examine PRP effect. According to PRP model, RT2 is lengthened at short SOA because operations requiring the bottleneck mechanism have to wait until this mechanism is finished with the first task. Thus, RT2 is essentially the sum of the time spent waiting for the bottleneck and the time needed for Task 2 processing.In PRP paradigm, the basic approach is to manipulate a factor that influences the difficulty of a specific second-task stage and then determine whether the effect of that factor decreases or remains constant as SOA decreases. If the factor affects a second—task stage at or beyond the bottleneck, the factor's effect should be additive with the effects of SOA (i. e. independent of SOA,). This is because SOA and the factor affect different additive components of RT2, SOA affects the time at which postbottleneck operations begin, whereas the factor affects the duration of postbottleneck operations. If the factor affects a stage prior to the bottleneck, however, the factor's effect should decrease with decreasing SOA (i. e., there should be an underadditive interaction of the factor with SOA). At short SOA, Task 2 operations that require the bottleneck (e. g., response selection) must wait until the bottleneck mechanism has completed Task 1 bottleneck processes. A factor affecting prebottleneck operations will only affect what goes on during this waiting time and therefore will have little or no effect on RT2, because the postbottleneck operations have to wait for the bottleneck mechanism to become available in either case. Of course, at long SOA the bottleneck mechanism will usually be available for Task 2 processing as soon as it is needed, and so the durations of prebottleneck operations will influence RT2. Thus, the factor will have a larger effect on RT2 at long SOA than at short one. Mental rotation have special character to use PRP effect research, Shepard and Metzler's classical research showed subjects pairs of perspective line drawings of choral shapes, and asked whether the shapes were identical or one was a mirror-image of the other. The figures in each pair were presented at different degrees of angular disparity, and the subjects' response times increased almost linearly as the angle between the figures increased. This means use mental rotation will central the difficulty of T2 more stably.In the present research, 14 reaction time experiments and 2 ERP experiments using a psychological refractory period paradigm examined whether the mental rotation is influenced by the response-selection bottleneck and whether the mental rotation process occurs in parallel with other cognitive operations. In each experiment, participants made speeded responses to both a tone (T1) and a different rotation letter or digit (T2), which presented with varying stimulus onset asynchronies (SOA). The results revealed that:(1) In the dual-task experiment, when mental rotation and other cognitive operations be presented in serially and quickly, T1 response selection influenced T2 response selection greatly, the effect of PRP was significant in RT2. The effect of mental rotation decreased substantially with SOA decreasing. The SOA between the two tasks is a major source in dual-task performance decrease. The shorter SOA, the worse dual-task performance. The tasks are more difficult, the more resource competition is in dual-task, which results worse the dual-task performance. PRP effect suggests when two highly overlapping tasks competed the limited capacity which results the T2 is influenced by the bottleneck.(2) In the highly overlapping dual-task, when the T2 is difficult, there is also a significant effect of SOA was observed in RT1. As SOA decreased, RT1 increased. T1 was significantly influenced by the difficulty of T2. This result showed that when T1 occupied the bottleneck to accomplish its response selection, mental rotation can parallel with other cognitive operation. This result supports the prediction of CCS model.(3) In the condition of Mirror-normal judgment, T2 difficulty effect still exists, but reduce slightly when the SOA is short. This effect does not wash out even when the SOA is 0 millisecond. But in the condition of classified judgment, the effect of mirror-normal and orientation is washed out, at this time, different orientation is identical to the perception judgment and lose it's mental rotation image character.(4) When subjects are required to respond to two stimuli presented in rapid succession, responses to the second stimulus are delayed. Such dual—task interference has been attributed to a fundamental processing bottleneck preventing simultaneous processing on both tasks. Two experiments show dual-task interference even when the first task does not require a response. The observed interference is caused by a bottleneck in central cognitive processing, rather than in response initiation or execution.(5) In the PRP paradigm, when the T2 is mental rotation image, the effect of orientation decrease substantially with decreasing SOA, especially when SOA is sufficiently short and make the dual-task's central overlapping.(6) In the highly overlapping dual-task paradigm, the main reason of dual tasks interfere with each other is both task competing for the limited central capacity at the same time. But the relationship between task-switching costs is the other main reason. The present experiments showed that when the two tasks in the same channel, the PRP effect also exists. These costs are traditionally attributed to fixed and unique capacity limitations for task set reconfiguration, target identification, and response selection, respectively.(7) The ERP finding showed that when SOA was short, the primary interference effect occurring at somewhat late stage of processing. These results generally support the model of CCS, which predicted that human being can perform tasks in parallel. The present research showed, even if does not response to T1, the PRP effect is still exist in RT2 in highly overlapping dual-task, this means the PRP effect is very robust and results performance decrease in dual-task paradigm.