Generation of random sequences by human subjects: Cognitive operations or psychological process?

Human subjects have difficulty in generating sequences that satisfy accepted requirements for randomness. This is often attributed to faulty cognitive operations or concepts of randomness. This article explores the alternative possibility that these deficiencies result from the operation of the same basic mechanisms for selecting responses that are studied in psychophysics. The features of random generation are briefly reviewed, an experiment is reported, and a model for random number generation is presented. The model assumes that an internal analogue source produces a random variable which may be represented on an internal dimension, the aleatory decision axis. The subject uses the measures produced by this random generator to select discrete responses. The mechanisms that select these responses use decision criteria whose positions on a decision axis are determined in accordance with a theory of the setting and maintenance of criteria presented elsewhere (Treisman & Williams, 1984). It is shown that this model can explain some important features of random sequence generation, and that what may be described as "cognitive intrusions" can be distinguished from effects of the response selection system. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Variations on the theme of feature integration: Reply to Navon (1990).

Feature-integration theory is outlined to clarify its claims and assumptions. The concepts of separated vs. integrated and of latent vs. manifest features are then examined. It is argued that quadratic polynomial search functions would be predicted only if the feature and the location were both latent—an unlikely assumption, against which there is considerable evidence from the efficiency of spatial cues to attention. An experiment on iconic memory suggests that attention can address features through their locations both for a physically present stimulus and for briefly stored sensory information. Finally, a comparison of D. Navon's (see record 1990-27540-001) preferred model to feature-integration theory seems to reveal no substantive differences. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Spreading suppression or feature integration? A reply to Duncan and Humphreys (1992).

J. Duncan and G. Humphreys (see record 1992-33782-001) propose an account of conjunction search in terms of interference from similar distractors, both through spreading suppression and through competition for input-template matching. Their reinterpretation of the calibration studies used to equate similarity by A. Treisman (1991) is questioned here. Furthermore, it is suggested that Duncan and Humphrey's new test, designed to equate similarity by matching the number of shared features, may actually require counting or conjoining of features in feature as well as in conjunction search. Another experiment is reported, which casts doubt on the idea that spreading suppression plays a major role in conjunction search. Finally, it is agreed that there is substantial convergence between the respective theories, but it still appears that conjoining features pose a special problem that cannot be explained solely by the grouping and matching mechanisms of Duncan and Humphreys. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Perceiving and re-perceiving objects.

A number of experiments exploring priming effects and automatization in the perception of novel objects are described, and a framework for understanding the benefits and costs of re-perceiving previously seen objects is proposed. The suggestion is that perceiving an object creates a temporary representation in an object file that collects, integrates, and updates information about its current characteristics. The contents of an object file may be stored as an object token and retrieved next time the object appears. This facilitates its re-perception when all of the attributes match and may impair it if some are changed. Thus, the world molds the mind to capitalize on earlier experiences but at the same time leaves it able readily to detect and represent any novel or unexpected objects and events. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Can decision criteria interchange locations? Some positive evidence.

In its simplest form, signal detection theory posits a single, invariant criterion for the yes/no detection task. But if paradigms with multiple criteria, such as the rating procedure, admit criterion variance, a new problem arises. If there are 2 or more response criteria on the decision axis and if these are each subject to random variation, the question arises as to whether a pair of criteria is constrained to maintain its nominal order at all times, or cause 2 criteria to change places. The theory of criterion setting developed by the 1st author and T. C. Williams (see record 1984-08523-001) is described, and it is shown that criterion-setting processes may produce criterion variance that is at least partly uncorrelated. Two alternative models for the organization of criteria in the rating procedure, the standard model and the distributed criterion model, are discussed. A prediction is derived that allows a test between the hypotheses that criteria are constrained to maintain a fixed order or that they may interchange locations on the decision axis. The results of 2 experiments, with 7 22–34 yr old observers (including the 2nd author), support the hypothesis that decision criteria may change places with one another. Alternative explanations are also considered. (20 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Search asymmetry: A diagnostic for preattentive processing of separable features.

Five experiments, with 42 university students, showed that the search rate for a target among distractors varied dramatically depending on which stimulus played the role of target and which that of distractors. For example, the time required to find a circle distinguished by an intersecting line was independent of the number of regular circles in the display, whereas the time to find a regular circle among circles with lines increased linearly with the number of distractors. The pattern of performance suggests parallel processing when the target had a unique distinguishing feature and serial self-terminating search when the target was distinguished only by the absence of a feature that was present in all the distractors. Results are consistent with A. Treisman and G. Gelade's (see record 1980-04685-001) feature-integration theory, which predicts that a single feature should be detected by the mere presence of activity in the relevant feature map, whereas tasks that require Ss to locate multiple instances of a feature demand focused attention. Search asymmetries may therefore offer a new diagnostic to identify the primitive features of early vision. Several candidate features are examined. Colors, line ends or terminators, and closure (in the sense of a partly or wholly enclosed area) appear to be functional features; connectedness, intactness (absence of an intersecting line), and acute angles do not. (40 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

A theory of criterion setting: An alternative to the attention band and response ratio hypotheses in magnitude estimation and cross-modality matching.

Discusses the attention band and response ratio hypotheses of D. M. Green and R. D. Luce (1974) and Luce and Green (1974) and notes some difficulties. An alternative hypothesis is put forward, based on a Thurstonian model for magnitude estimation in which the presented stimulus intensities are subjected to a logarithmic transformation. Response criteria are then applied to the resulting quantities to select corresponding responses. The setting and maintenance of these response criteria are accounted for by a theory of criterion setting previously developed by the present author and T. C. Williams (1984). A similar model is developed for cross-modality matching, and it is shown that these models can predict the V pattern for the coefficient of variation of response ratios, can predict the inverted V pattern for correlations between successive responses, and can account for some of the difficulties found in the literature. (39 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Relation between signal detectability theory and the traditional procedures for measuring sensory thresholds: Estimating d' from results given by the method of constant simuli.

The theory of signal detectability assumes that the central effect of a stimulus varies because of physical and neural noise; consequently, the detection of a signal requires a central statistical decision procedure. Similar assumptions have been made by psychophysicists to explain the results of traditional threshold measurement procedures. The interrelations between signal detectability and threshold measures are discussed in relation to psychophysical statistical decision theory, and it is shown that (a) the false positive rate should be related to the Crozier ratio C = ΔΙ/?ΔΙ, and (b) it should be possible to use responses given in the method of constant stimuli to predict the value of d' that will be assigned to a given stimulus by a signal detectability procedure. Evidence supporting both predictions is reported, and the relation between threshold measures and "personality tests" is discussed. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Illusory words: The roles of attention and of top–down constraints in conjoining letters to form words.

Conducted 3 experiments with 40 undergraduates to confirm that illusory recombinations of letters can be seen when several words or letter-strings are presented too briefly to allow focused attention to each item in turn and to explore the role of lexical entries in promoting or preventing these preceptual errors. Overall findings indicate that when attention was divided among 4 briefly exposed syllables, Ss mistakenly detected targets whose letters were present in the display but in the wrong combinations. These illusory conjunctions were somewhat more frequent when the target was a word and when the distractors were nonwords, but the effects of lexical status were small and did not reach significance in free report of the same displays. Search performance was further impaired if the nonwords were unpronouncable consonant strings rather than increased conjunction errors. Results are discussed in relation to feature-integration theory and to current models of word perception. (46 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)