Stimulus Set Size Research Articles

Normative and Descriptive Analyses of Simpson's Paradox in Decision Making

Suppose that you are evaluating two delivery companies. Your investigation shows that Company B has a better on-time rate for small packages and also for large packages. Despite Company B's performance, however, Company A has a better overall on-time rate. This situation exemplifies Simpson's Paradox, in which the judged relationship between two variables (e.g., company and performance) differs depending on whether that relationship is viewed within subcategories of a third variable (e.g., package size) or in the aggregate. A normative analysis is presented arguing that the reasonableness of using the third variable depends upon the sample size as well as the separation between and variability within categories. To test subjects' abilities to behave appropriately in Simpson's Paradox situations, we examined responses to variations in these factors in five studies. Results showed that subjects had little or no sensitivity to differing stimulus set sizes. Also, subjects were sensitive to relationship strength and the variability within and between groups, and in nonnormative ways. Subjects' judgment behavior is related to a broader perspective concerning selection among multiple available levels of analysis based on a consideration of argument strength.

The effects of stimulus set size and word frequency on verbal serial recall

Two experiments are reported which examine immediate serial recall for high-and low-frequency words. The words in each list were either repeatedly drawn from the same small pool of candidates (in the closed set conditions) or each word only ever occurred once during the experiment (in the open set conditions). The results consistently show an effect of word frequency but the effect of set size was only apparent for low-frequency words. It is argued that both frequency and set size effects reflect processes concerning the “clean-up” of degraded short-term memory traces.

The roles of stimulus and response set size in the identification and categorisation of unidimensional stimuli

Dual mapping theories (e.g., Atkinson, Bower, & Crothers, 1965; Lacouture & Marley, 1991, 1995) propose that psychophysical judgment involves two component processes: stimulus representation and response selection. This empirical and theoretical paper studies the relative contributions of these two component processes in determining performance in the identification and categorisation of unidimensional stimuli. We extend Lacouture and Marley's (1995) mapping model of absolute identification to categorisation, and demonstrate that the extension predicts that the number of response categories has a much larger effect than the number of stimuli in determining various aspects of the relevant data, including the so-called bow effect. This prediction is supported by the results of three experiments in which the number of stimuli and/or the number of response categories was manipulated.

Reprogramming phonologically similar utterances: the role of phonetic features in pre-motor encoding.

The effects of phonologic similarity on speech production latencies were investigated to explore the role of articulatory phonetic features and reprogramming operations during pre-motor stages of production. A form-based priming technique was used in five experiments to elicit rapid productions of single words. Subjects responded to visually presented pairs of minimally contrastive monosyllabic words, which varied with respect to the phonetic featural similarity of their onsets. Factors that were systematically manipulated across experiments included stimulus set size, duration of the interstimulus interval, and the use of feedback to subjects concerning their response latencies. Speech onset latencies obtained in a control condition in which no phonetic features were shared were compared to four other conditions in which the word initial phonemes of prime-target pairs did share features. Results revealed that shared manner was the most influential factor associated with the observed inhibitory phonologic similarity effect. In addition, smaller stimulus set size (6 words) yielded significantly slower overall response latencies than experiments employing larger stimulus sets (18 words). These findings suggested that inhibitory phonologic similarity effects did not stem from biomechanical constraints imposed by the articulatory system. Rather the methods employed in this investigation were supported as a means to investigate both the underlying units of representation and the processes involved in pre-motor planning apart from articulatory effects. The results of this investigation also supported the hypothesis that, during phonologic encoding, word form retrieval entails the selection and assembly of sublexical units into word form frames. No evidence of whole word retrieval during pre-motor encoding was obtained. The potential utility of this experimental paradigm in the investigation of pre-motor planning in disorders that putatively affect the processes involved in transforming word meaning into word form is also discussed.

Allocation of attentional resource and monitoring processes under rapid serial visual presentation

With the use of rapid serial visual presentation (RSVP), the present study investigated the cause of target intrusion errors and functioning of monitoring processes. Eighteen students participated in Experiment 1, and 24 in Experiment 2. In Experiment 1, different target intrusion errors were found depending on different kinds of letters --romaji, hiragana, and kanji. In Experiment 2, stimulus set size and context information were manipulated in an attempt to explore the cause of post-target intrusion errors. Results showed that as stimulus set size increased, the post-target intrusion errors also increased, but contextual information did not affect the errors. Results concerning mean report probability indicated that increased allocation of attentional resource to response-defining dimension was the cause of the errors. In addition, results concerning confidence rating showed that monitoring of temporal and contextual information was extremely accurate, but it was not so for stimulus information. These results suggest that attentional resource is different from monitoring resource.

Procedural learning: I. Locus of practice effects in speeded choice tasks.

When subjects perform 750 trials of a speeded choice task mapping different categories of symbols (e.g., letters, digits) onto different responses, excellent transfer is observed to new items in the trained categories (Experiment 1). However, when arbitrary sets of stimuli are mapped onto each response, introducing new stimuli substantially retards performance (Experiment 2), even when the size of the potential stimulus set remains constant (Experiment 3). Surprisingly, responses to already trained items are as slow as responses to new items in these transfer tests. When the mapping is categorical, shuffling the assignment of stimuli to responses drastically slows responses (Experiment 4). However, changing to a spatially homologous mapping with responses on the other hand produces excellent transfer (Experiment 5). Together, these results indicate that practice in speeded choice tasks affects primarily the response selection stage, rather than perceptual processing or motor responses. These data suggest that practice primarily strengthens links between category representations and spatially defined responses. Furthermore, when an arbitrary collection of symbols is mapped onto a given response, practice produces an ad hoc category representation and strengthens links between individual items and the category, as well as links between the category and the response. With practice, people become dramatically faster and more accurate in sensorimotor skills. Characterizing these changes is of major importance, for both applied and theoretical purposes. In applied situations, it would be useful to know how conditions of skill acquisition can be structured so that learning is most efficient and to know when one form of training will produce useful transfer to other, related tasks. For theoretical purposes, understanding the functional character of skill learning should provide a basis for developing detailed models of the changes that underlie such learning (such as neural net models). The present article focuses on the changes in performance that occur with modest amounts of practice in very simple sensorimotor tasks, namely those requiring speeded choice responses. In speeded choice tasks, subjects are presented with a single stimulus and must rapidly select and execute a response depending on the identity of the stimulus. This is one of the simplest paradigms in which one can examine procedural skills and their development, and it provides a useful laboratory model for the numerous human skills that involve retrieving and producing arbitrary responses to stimuli. Performance in speeded choice tasks is subject to various kinds of change as subjects perform repeated trials. Subjects' responses become faster (and often more accurate)—the prac

Brain potentials in a memory-scanning task. II. Effects of aging on potentials to the probes

Brain potentials accompanying the classification of probe items as being members of previously presented list were recorded from subjects ranging in age from 18 to 86 years old. A group of older subjects (averave age = 66 years) was compared to a younger group (average age = 29 years). The items tested were verbal (digits) and non-verbal (musical notes). Digits were presented in the auditory and visual modalities, and notes were presented acoustically. Reaction times (RTs) and performance accuracy were computed. Potentials are described in terms of scalp distribution, latency and amplitude as a function of the type of stimulus (verbal/non-verbal, auditory/visual) and age group (younger/older). Evoked potentials to target notes in an auditory target-detection (‘odd-ball’) task were also recorded for comparison with the memory tasks. Potentials evoked by probes consisted of a sequence of sensory components in the first 250 msec followed by a cognitive component that was positive in polarity and sustained in duration (approximately 700 msec labeled P3), consisting of an earlier frontal component, P3a (mean latency: younger = 385 msec, older = 406 msec), and a large (15 μV) and later parietal constituent, P3b (mean latency: younger = 574 msec, older = 630 msec). The frontal derivation of the younger subjects showed a sustained negative bias of the wave forms in the latency range of 200–500 msec (P2 to P3) compared to the older subjects. Reaction times were longer in older subjects than in younger subjects for all stimulus types and set sizes. For the potentials evoked by the probes the younger group had consistently larger late parietal components (P3b) than the older group, whereas the late frontal potentials (P3a) were larger for the older than younger subjects. Except for visual stimuli, the latencies of the parietal sustained potentials were not influenced by subject age in contrast to the uniform changes in RT for all stimulus types. Significant amplitude and latency effects on the parietal sustained potentials accompanied the different stimulus types and memorized-set sizes which were similar for the two age groups. These results suggest that the effects of aging on short-term memory are primarily on response selection, as evidenced by RT slowing with aging, and not on memory-scanning processes as evidenced by the similarity of the latency measures of the accompanying brain potentials between the two age groups.

Acceptability of text‐to‐speech systems: Its quantification using magnitude estimtion and its relationship to intelligibility and naturalness in various degrees of distortion

Because there are no physical measurements that quantify perceptual attributes of synthesized speech, subjective tests must be used. In this project the possibility of directly scaling acceptability via magnitude estimation is assessed. One hundred and twenty subjects took part in the study. Seven different synthesizers and three types of background noise were employed. The results are discussed in light of the following questions. (1) What is the relationship between acceptability on one hand and naturalness and intelligibility on the other in various degrees of distortion? (2) Are the objective intelligibility scores and subjective magnitude estimations highly correlated in all conditions? (3) Do the relative ratings produced by different groups of subjects agree? (4) Are the ratings on an absolute scale? (5) Are the ratings invariant to the stimulus set and range size? (6) Are the ratings practice invariant? (7) Do the ratings depend on the subject's familiarity with the test material?

Categorization of natural stimuli by monkeys (Macaca mulatta): Effects of stimulus set size and modification of exemplars.

Categorization of natural stimuli by monkeys (Macaca mulatta): Effects of stimulus set size and modification of exemplars.

Categorization of natural stimuli by monkeys (Macaca mulatta): Effects of stimulus set size and modification of exemplars.

The concept humans was studied in two experiments on rhesus monkeys, in each of which a two-choice simultaneous discrimination procedure was used. In Experiment 1, the choice was between scenes with humans and scenes without humans, with the slide set sizable enough that a large number of trials could be given without repeating any individual slide. Speed of categorization learning was faster and final level of performance was higher than in prior research in this laboratory involving a much smaller slide set. Experiment 2 was an attempt to obtain some information about the basis for the categorization by means of a series of probe trials. Probe trials involved slides of humans that were modified in one of several ways and slides in which monkeys or apes were present instead of humans. When paired with slides with humans, probe slides were seldom chosen, except when they showed a human rightside up in an upside down scene. In the latter case, choices were at the chance level. When paired with a slide with no humans in the scene, probe slides were usually chosen, except when they showed monkeys or apes or silhouettes of humans, in which case choices were again at the chance level. Possible reasons for the differences in results of category learning tests with pigeons and monkeys are discussed as are the implications of the probe tests for a concept interpretation of these results.

Jensen's reaction-time studies: A reply to Longstreth

Longstreth's (1984) critique of Jensen's research on the relationship of IQ to individual differences in visual reaction time (RT), measured in the Hick paradigm, is found to have numerous errors of fact and interpretation, some trivial and some of theoretical importance. Longstreth's narrowly focused and conjectural style of criticism, which peculiarly strains to favor the null hypothesis, unfortunately obscures the essential findings of Jensen's (and others') studies of the RT-IQ relationship. The two main negative verdicts of Longstreth's critique concerning the RT-IQ relationship are refuted by meta-analyses of presently available data. First, not only do individual differences in RT show a significant negative correlation with IQ, but individual differences in the slope of the regression of RT on stimulus set size scaled in bits (i.e., the binary logarithm of the number of potential reaction stimuli) also show a fully significant, albeit low, negative correlation with IQ. Contrary to Longstreth's second negative surmise, meta-analysis also shows that the magnitude of the RT-IQ correlation itself is a linearly increasing (negative) function of stimulus set-size scaled in bits.

Probing Proctor's priming principle: the effect of simultaneous and sequential presentation on same-different judgments.

Three explanations (internal noise, priming, relative frequency) of the fast-same effect were examined. The internal-noise principle, which predicts more errors as well as faster correct judgments on same pairs, was consistently confirmed even with sequential presentation. More elusive was priming, that is, the facilitation of encoding from letter repetition with sequential presentation. Priming was inhibited by the presence of intertrial letter repetition. The error data indicated that priming involves an increased efficiency in encoding (d'), as Proctor claimed, rather than a criterion shift (beta), as Krueger and Shapiro claimed. An alternative to the priming explanation, based on the greater susceptibility of simultaneous presentation to analytical processing, was tested and disconfirmed. Stimulus-set size did not affect the speed advantage for same pairs, thus disconfirming the relative-frequency explanation, according to which same judgments are faster because typically there are fewer unique same than different pairs.

Different difficulty manipulations interact differently with task emphasis: evidence for multiple resources.

To test the notion of multiple resources, a two-dimensional pursuit tracking task was paired with a letter-typing task, the difficulty of which was manipulated by varying cognitive (size of stimulus set) and motor (repetitiveness of finger chords) factors. In addition, task priority was manipulated. The latter factor had a large effect on the performance of the two tasks, which indicates that they compete for resources. Both types of typing difficulty manipulations affected typing performance, but only motor difficulty interacted with priorities. Since difficulty manipulations that tap resources common to both tasks are predicted to interact with priorities, the results are interpreted to indicate that in joint performance, typing and tracking compete mainly for motor-related concept, the letter-typing task is argued to require at least two kinds of resources.

Isolating the Typicality Effect in Semantic Memory

An experiment was conducted to isolate the typicality effect in the overall categorization process. Subjects were required to indicate whether or not a probe word was a member of one of a variable number of categories held in memory. The typicality of the probe word, with respect to its superordinate category, was varied as was the physical quality of the probe. These manipulations were designed to test predictions drawn from Collins and Loftus (1975) and Anderson and Reder (1974), as well as a prediction based on the well-known effect of semantic context on word perception. A fourth prediction was drawn from a proposed model which postulates successive access of categories and locates the typicality effect in a within-category search stage. Typicality proved to be additive with stimulus quality and set size, but interacted with response type. These results were interpreted as support for the successive-access within-category search model while disconfirming the alternative predictions.

Effects of time stress and stimulus-response set size on the efficiency of detection of involvement with suppressed information through the use of the forced number-guessing technique

Effects of time stress and limitation of stimulus and response set size on the detection of suppressed involvement with information through the forced number-guessing technique, were investigated. Variations in load do not affect the overall detection efficiency, but the mechanisms through which ‘guilty’ Ss expose themselves are a function of the specific loading conditions applied. The results indicate that information suppression might be considered as an information processing paradigm.

SPEED AND ACCURACY OF ABSOLUTE PITCH JUDGMENTS: SOME LATTER‐DAY RESULTS

ABSTRACTNine subjects, 5 of whom claimed absolute pitch (AP) ability (4 from childhood, 1 by self‐training) were given a pitch judgment task in which they had to strike notes on the piano as rapidly as possible to match randomized tape‐recorded piano notes. Stimulus set sizes were 64, 16, or 4 consecutive semitones, or 7 diatonic notes of a designated octave. A control task involved motor movements to notes announced in advance, with the effect that set size was 1. Accuracy, measured on the basis of deviations of responses from targets, significantly differentiated AP from NAP (non‐AP) subjects at all set sizes except the control task. For both groups, accuracy increased as set size decreased. Decision times, measured as that part of total response time before a movement to the response note began, decreased as set size decreased, but did not differentiate AP and NAP subjects. The performance of the trained AP subject was not distinguishable from that of the remaining AP subjects either in accuracy or decision time. Results are discussed in terms of a two‐factor theory of AP ability whereby NAP subjects use only relative tone height as the basis of judgment but AP subjects also use standards of tone chroma stored in long‐term memory. The abnormally high channel capacities and rates of information gain for AP subjects are interpreted on the basis of transmission of information in two channels, whereas NAP subjects transmit information in only one channel.

Time and Accuracy Measures for Alternative Multidimensional Scaling Data Collection Methods

The impact of stimulus set size and data collection methods on response accuracy and response time were investigated in an empirical study. Anchor Point collection techniques proved superior in reducing respondent time with no sacrifice in accuracy.

Stimulus Dimension Pre-Training and Set Size in Learning Multiple Discriminations with Letters of the Alphabet1

A variety of tasks in early education such as naming letters and shapes may be adapted to the paradigm for learning multiple discriminations and thus may share common elements which facilitate learning. It was hypothesized that acquisition of a multiple discrimination task involving letters of the alphabet would be facilitated by pre-training in a relevant as contrasted with an irrelevant stimulus dimension, by reducing the number of letters initially presented in a set, and by arranging the letters in the reduced sets so they may be cumulatively reviewed. Children aged 4-6 generally learned more effectively when they were pre-trained on a relevant dimension and allowed to cumulatively review smaller sets as they learned the full list.

Processing time for naming and categorization of letters and numbers

A reaction-time experiment was carried out to examine the relationship between naming and categorization. Ss were shown one item at a time and asked either to name the item or to categorize the item as a letter or number. The size of the stimulus set was varied systematically across Ss. Naming time increased as the stimulus set size increased; categorization time could be predicted by the time required for naming, plus a constant. These results were interpreted as indicating that naming must precede categorization.

Efficiency of GSR detection of information as a function of stimulus set size.

ABSTRACTThe reported studies attempted to evaluate the influence of stimulus set size on the efficiency of detection of information through the analysis of GSR responses. Three stimulus set size variations of a standard card detection test were employed. No significant reduction in absolute detection scores was found as the number of cards was increased, and an analysis based on the Theory of Signal Detection also suggested better autonomic discrimination as the stimulus set size increased.The results suggest that the subject responds to the stimuli by dividing them into a single relevant stimulus and a reject category containing all others.