Multidimensional scaling versus components analysis of test intercorrelations.

Considers the relation between coordinate estimates in components analysis and multidimensional scaling (MDS) and reviews algebraic relations between metric MDS and components analysis. Previous studies suggest that the same relations usually characterize components and nonmetric MDS analyses of correlation matrices. Merits of the MDS and factor approaches are compared, and results are related to other methodological issues surrounding research on the general ability factor, response tendencies in self-ratings, and halo in employee evaluations. (53 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Multidimensional scaling of the causes for negative interpersonal behavior.

Conducted studies to determine empirically the dimensions underlying attributions made for negative behavior in a close interpersonal relationship (a marriage). 96 university students rated the similarity of 13 causes given either by the enactor of a negative behavior (i.e., the actor) or by the person affected by the behavior (the partner or spouse). A multidimensional scaling analysis of these data yielded 2 dimensions in each condition. The interpretation of these dimensions was guided by data from 68 additional Ss who rated the 13 causes on bipolar scales (e.g., temporary–permanent). Dimension 1 in both conditions was interpreted as "positive vs negative attitude toward spouse." Dimension 2 of the actor condition was interpreted as "intentional vs unintentional" and Dimension 2 of the partner condition as "actor's traits vs circumstances or states." These dimensions are discussed in relation to the a priori causal distinctions made in current attribution writings and previous findings concerning actor–observer biases. (18 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Multidimensional scaling analyses of the perceived social structure of informal groups.

Four groups of student hallmates provided similarity ratings of each other, and they rated each other on various characteristics. Group-level multidimensional scaling analyses suggested that there was no consensual social structure in any of the four groups. In contrast, individual-level analyses found that individuals had coherent perceptions of the social structures of their groups, although no 2 members of any group perceived the social structure of their group similarly. Nevertheless, there was considerable similarity among people in the dimensions underlying their perceived structures. Conventionality, friendliness, influence, and interpersonal openness served as organizing dimensions for the majority of participants, and sense of humor, how hard it was to get along with people, introversion, interesting, and academic orientation were important bases for about a third of the participants. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Dr. Jekyll and Mr. High: Multidimensional scaling of alcoholics' self-evaluations.

Developed a social cognition paradigm to study 40 male alcoholics' self-evaluations. Ss made judgments of similarity among all possible pairs of a set of 15 stimulus people. The "target" stimuli in this set were sober self and high self. The "marker" stimuli included a broad array of hypothetical people representing various personality types. Data were scaled according to multidimensional successive intervals. Results suggest that alcoholics consider themselves to be entirely different people when sober than when high. The stability and meaningfulness of these findings support the continued development of this assessment procedure for further investigation in the alcoholism area and also for general personality research. (35 ref.) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Multidimensional scaling of emotional facial expressions: Similarity from preschoolers to adults.

Structural models of emotion represent the fact that emotions are perceived as systematically interrelated. These interrelations may reveal a basic property of the human conception of emotions, or they may represent an artifact that is due to semantic relations learned along with the emotion lexicon. The 1st alternative was supported by results from a series of scalings of 20 emotional facial expressions, results that could not easily be attributed to word similarity. Similarity data on the facial expressions were obtained from 30 undergraduates and 42 4–5 yr olds. For both groups, similarity was measured without the use of emotion labels by asking Ss to group together people who appear to feel alike. The structure of emotions obtained from both children and adults was as predicted: a roughly circular order in a 2-dimensional space, the axes of which could be interpreted as pleasure–displeasure and arousal–sleepiness. The form and meaning of this structure was supported through 2 additional scalings of the facial expressions with adults: a multidimensional scaling based on direct ratings of similarity–dissimilarity and unidimensional scalings on the pleasure–displeasure and arousal–sleepiness dimensions. (27 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Multidimensional scaling of color patterns of the DIN color chart

The DIN color chart was developed in the 1950s by Manfred Richter using of classical psychophysical scaling techniques. It is based upon the idea that colors are ordered along three subjective dimensions, i.e. hue, saturation, and brightness. Furthermore, it is assumed that the colors of the DIN color chart fulfill the principle of specific equidistancy. The main aim of this study was to investigate this claim empirically. More specifically, it was tested whether colors of the same hue and brightness are equally spaced along the dimension of subjective saturation. The data were collected in three paired comparisons experiments and were analyzed using multidimensional scaling. According to the results, the psychological properties of the DIN color system can be replicated approximately. In addition to a confirmation of the postulate of equidistancy this research contributes to the understanding of the convergence of different scaling methods.     

Multidimensional scaling of synthetic musical timbre: Perception of spectral and temporal characteristics.

Investigated the perceptual correlates of acoustic parameters involved in musical timbre by examining judgements of timbre dissimilarity. Nine synthetic sounds were created, derived from crossing 3 levels of spectral and temporal parameters (number of harmonics and rise time). Two separate conditions were tested, 1 using single tones, the other using short melodies. 15 musically untrained 20–47 yr olds were presented with pairs of stimuli and judged dissimilarity on an 8-point scale. The spatial configuration resulting from multidimensional analysis of the data was best fit by a 3 dimensional model, with the 1st 2 dimensions accounting for most of the variance. The perceptual space derived from the analysis indicates that these 2 orthogonal dimensions corresponded closely to the spectral and temporal differences inherent to the stimuli. Similar results were obtained with both melodies and single tones. A 2nd experiment replicated the findings despite the introduction of random loudness variation. It is concluded that even musically unselected subjects are sensitive to spectral and temporal information in musical tones, and are able to use them independently in making perceptual judgements of musical timbre. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Multidimensional scaling of complex sounds by school-aged children and adults

A paired-comparisons procedure was used to evaluate the processing of complex, nonspeech sounds by 7- and 10-year-old children and adults. Stimuli were brief duration and included pure tones, harmonic complexes, and bands of noise. From their similarity ratings, a three-dimensional multidimensional scaling solution was derived. Results suggested that listeners classified the stimuli into clusters based upon periodicity and the number of spectral peaks. Within each cluster, the stimuli were ordered according to frequency. Because individual differences in the overall weightings of features were large, separate solutions were derived for two subgroups of listeners, formed based upon similarities in the pattern of dimension weights obtained in the group analysis. One subgroup, for whom the full group analysis captured a large proportion of the variance in the ratings, included the adults, many of the 10-year-olds, and a few of the 7-year-olds. The solution derived for this subgroup suggested that spectral and temporal information were weighted equally and integrated into all dimension weights. Frequency information was coded but given less weight and was not used for stimulus classification. A second subgroup of listeners included most of the 7-year-old and some of the 10-year-old children. Examination of their data suggested that they relied heavily on an analysis of the signals according to periodicity as was reflected in their temporal fine structure. Also encoded but of lesser importance were aspects of spectral shape and absolute frequency.     

Self-regulatory effects of monitoring sensory and affective dimensions of pain.

Examined the differential effectiveness of focused attention on either affective discomfort or sensory components of electric-shock-induced pain experiences in 30 undergraduates. In addition, the impact of models displaying tolerance or intolerance for pain was examined on reports of these different components of the experience. Attention to sensory components led to lower pain tolerance, avoidance of shocks at lower current levels, and characterizations of the shocks as more painful than stronger current intensities accepted during the discomfort rating task. The strategy of modeling tolerance was effective in influencing both sensory and affective reports of the experience. (PsycINFO Database Record (c) 2010 APA, all rights reserved)