Serial Non-Randomness in Auditory Differential-Thresholds as a Function of Interstimulus Interval

Day, Willard F.

doi:10.2307/1419040

Cited by 21 publications

(6 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Later McGill (1957) reported avoidance of repetition of responses with subthreshold stimuli in a four-alternative signal detection task. Reviewers of work in psychophysics have noted the nonrandom patterning of responses with subthreshold stimuli, although not all have mentioned a tendency for alternations to be predominate (Day, 1956, 1957; Senders, 1953; Senders & Sowards, 1952). Tune (1964b) described Fernberger's and related early work on the “contrast effect” as “the historical root of much of the contemporary speculation on the stochastic nature of behavior” (p. 286).…”

Section: The Production Of Randomnessmentioning

confidence: 99%

The production and perception of randomness.

Nickerson¹

2002

Psychological Review

205

200

View full text Add to dashboard Cite

This article contains a discussion of the elusive nature of the concept of randomness, a review of findings from experiments with randomness production and randomness perception tasks, and a presentation of theoretical treatments of people's randomization capabilities and limitations. The importance of task instructions and the difficulty of interpreting results when instructions are vague or ambiguous are stressed. The widely held view that people are incapable of generating or recognizing randomness is shown to lack the strong experimental support that has sometimes been claimed for it.

show abstract

Section: The Production Of Randomnessmentioning

confidence: 99%

The production and perception of randomness.

Nickerson¹

2002

Psychological Review

205

200

View full text Add to dashboard Cite

show abstract

“…255), he did not include some relevant work done before 1964. For instance, in psychophysics long intertrial intervals have systematically been used to reduce the sequential nonrandomness in subjects' responses to threshold stimulation (63,69,88). Moreover, a number of researchers explicitly investigated the effect of speed on repetition avoidance in free association (148,241,247).…”

Section: Effect Of An Increared Number Of Items From Which Selectionsmentioning

confidence: 99%

Variables That Influence the Generation of Random Sequences: An Update

Brugger¹

1997

Percept Mot Skills

114

125

View full text Add to dashboard Cite

This article updates Tune's 1964 review of variables influencing human subjects' attempts at generating random sequences of alternatives. It also covers aspects not included in the original review such as randomization behavior by patients with neurological and psychiatric disorders. Relevant work from animal research (spontaneous alternation paradigm) is considered as well. It is conjectured that Tune's explanation of sequential nonrandomness in terms of a limited capacity of short-term memory can no longer be maintained. Rather, interdependence among consecutive choices is considered a consequence of an organism's natural susceptibility to interference. Random generation is thus a complex action which demands complete suppression of any rule-governed behavior. It possibly relies on functions of the frontal lobes but cannot otherwise be "localized" to restricted regions of the brain. Possible developments in the field are briefly discussed, both with respect to basic experiments regarding the nature of behavioral nonrandomness and to potential applications of random-generation tasks.

show abstract

“…Negative bias is usually obtained, but one could make a strong a priori case for a positive recency effect (i.e., persistent repetition of a certain response). Wiegersma (1982b) calls such an explanation a “strength theory” because the strongest and most available response at each stage is the one just produced (see Day, 1957; Holland & Lockhead, 1968; Ross & Levy, 1958; Verplank & Blough, 1958; Ward & Lockhead, 1971; Weiss, 1964). Such a pattern of results would be equally consistent (or inconsistent) with the representativeness heuristic.…”

mentioning

confidence: 99%

A Markov model for generation of random binary sequences.

Budescu¹

1987

Journal of Experimental Psychology: Human Perception and Perfor

View full text Add to dashboard Cite

Numerous experiments have confirmed Reichenbach's (1934Reichenbach's ( /1949 expectation that humans are suboptimal generators and judges of randomness-they tend to generate series with too many alternations (the negative recency effect) and to match the probability characteristics of the process in short subsequences (the local representativeness effect). Several classes of theories have been proposed to explain these results, with modest success. In the present article a stochastic model assuming one free parameter of serial dependence is outlined, and an experimental procedure for its test is developed. In the first experiment the model's parameters are estimated and its major assumptions are successfully tested. Fifteen of the 18 subjects are well fitted by the model, but only 11 of them display negative bias. The second experiment tested and rejected an artifactual interpretation of the results based on Neuringer's (1986) suggestion that people can be trained to become good randomizers. The third experiment extended the validity of the model by demonstrating the stability and robustness of its parameters under systematic changes in the display of previous responses. Implications of the model and the data for the various theories are discussed. It is suggested that only theories which are based on the notion of a "subjective concept of randomness" are consistent with the results of these experiments.The research on subjective randomness has recently celebrated its 50th birthday. Its beginnings can be traced to a brief observation made by Reichenbach (1934Reichenbach ( /1949 in his book on the foundations of probability theory. He speculated that people untrained in probability would be unable to generate random sequences and would produce instead series with too many alternations. This observation was confirmed in a variety of experiments reviewed by Tune (1964aTune ( , 1964b and in a most detailed and systematic fashion by Wagenaar (1972aWagenaar ( , 1972b.The experimental work in this domain has used three paradigms. In randomization experiments, subjects were instructed to produce a sequence of responses with random characteristics, that is, "a random sequence." Such experiments were performed with various numbers of alternative responses (from binary series with two responses to 26 alternatives), with various lengths of sequences (from 20 to 2,520), and with a variety of stimuli (numbers, letters, symbols, etc.). Subjects responded by writing or calling out the sequence or by pushing buttons and stamping symbols at different rates of response. A detailed summary of all the experimental conditions and response procedures was provided by Wagenaar (1972aWagenaar ( , 1972b. In the constrained randomization procedure the subjects are presented with all the components of the sequence (e.g., 50 blue cards and 50 yellow ones) and are required to arrange them in random

show abstract

Serial Non-Randomness in Auditory Differential-Thresholds as a Function of Interstimulus Interval

Cited by 21 publications

References 0 publications

The production and perception of randomness.

The production and perception of randomness.

Variables That Influence the Generation of Random Sequences: An Update

A Markov model for generation of random binary sequences.

Contact Info

Product

Resources

About