Lens model

from Wikipedia, the free encyclopedia

The lens model (English lens model ) is a general conceptual framework of the sentencing process, the psychologist Egon Brunswik back (1903-1955). Organisms must continuously various situations and objects assess which are to be assessed properties not usually immediately perceptible (z. B. the calorie content of desserts), but from references (eng. Cues ) are opened (need such. As size, taste, Cream content of the dessert). The information usually does not allow a perfect conclusion about the property, but is often statistically related to it. A lens model analysis provides information on how the individual indications are integrated into an overall assessment. The basic idea as well as the analysis techniques associated with the lens model have had a lasting impact on empirical judgment and decision research since the 1950s and, in particular, enabled many insights into the acquisition of expertise and the conditions for learning good judgment strategies.

Schematic representation of the "lens model" according to Egon Brunswik

Basic idea of ​​the lens model

Since the property or dimension of a judgment object to be assessed - the judgment criterion - is usually not directly accessible to the sensory experience, it must be inferred from cues that are statistically related to the judgment dimension. The stronger the statistical connection between a reference and the criterion, the more useful it is as a basis for a judgment. The person making the judgment accordingly looks at the object of judgment, metaphorically speaking, through the "lens" of the available cues, from where the model got its name.

The approach is universally applicable: For example, we assess the probability of rain on the basis of many available information (e.g. weather report, cloud cover, temperature, air pressure) or the danger of a traffic situation on the basis of several situation variables (clarity of the intersection, speed of the vehicle). Likewise, a medical diagnosis is derived from various symptoms and laboratory or test results, etc. Various indications can conflict and suggest different judgments (e.g. cloudy sky but positive weather report).

Lens model analysis makes assumptions about how such information is integrated (and conflicting information weighed against each other). Using statistical methods, in particular multiple linear regression analysis , information about a person's psychological judgment process can be obtained for the experimenter, such as the accuracy of the judgments, the use of individual cues, the quality of the fit between the judge and the environment, and the consistency of the judgments.

History of origin

Brunswik developed the preliminary idea for the lens model analysis in investigations into perceptual judgments in connection with the phenomenon of size constancy . He developed the lens model as a general model of the judgment process in his book The conceptual framework of psychology , published in 1952 . His student Kenneth R. Hammond recognized that the framework model can be applied to arbitrary judgment situations, e.g. B. clinical judgments of psychologists and doctors, can be expanded. The concept and the analysis methods were further developed under the name "Social Judgment Theory".

Concept and basic terms

The criterion value Y E to be assessed (the index E stands for environment) of a distal object is unknown to the person making the judgment. Typically there are a number of cues X i that are statistically related to the criterion, e.g. As a specific symptom X , the probability of a particular diagnosis Y E increased. The strength of the statistical relationship between a reference and the criterion is referred to as its ecological validity and indicates how much this reference objectively contributes to the prediction of the criterion. The relation of a set of cues to the criterion is called the environmental model. In contrast, the judgment model on the other side of the "lens" describes how the indications X i influence the actual judgments Y S (index S for subject ). The strength of the statistical relationship between a note and the judgment is called Information on usage (Engl. Cue utilization ) indicates and measures how much a judgment Ender takes into account the respective Cue in the judgments.

As a measure of the judgment output (engl. Achievement ) is usually the product-moment correlation r a used between actual criterion values and judgments. This mainly depends on (1.) the ecological validities, (2.) the consistency of the judgment and environmental model and (3.) the consistency of the judge: The level of the ecological validity indicates how precise the criterion is generally based on the information can be determined and how much the individual cues contribute to the prediction. The correspondence between the judgment model and the environmental model is high if valid information is used accordingly by the judge, while less valid information is given less weight. There is a high level of consistency when the person making the judgment constantly assesses the same objects of judgment in multiple assessments and shows no random fluctuations in the judgment.

Formalization and analysis

A lens model analysis can be applied to a series of judgments about objects when the true criterion values Y E , the judgments Y S and the values ​​of the cues X i are present. The environmental and the judgment model are each determined by a multiple linear regression of the Y E and Y S on the values ​​of the indications as predictors . The regressions provide multiple correlation coefficients R E and R S as measures of the respective model quality and regression weights B i (or β i ) as measures of the ecological validities (environmental model) or the reference uses (judgment model). In addition to a qualitative comparison of the regression weights of both models, a formal analysis using the lens model equation is possible, which is mostly used today in the form formulated by LR Tucker. The lens model equation is:

The judgment performance r a is therefore composed of the fit index G , the multiple correlations R E and R S , as well as a technical term C , which is calculated from the correlation of the regression residuals. The index of fit G measures the correspondence between the environmental model and the judgment model and therefore represents a measure of the suitability of the judge to the judgment environment. Since R E reflects the linear predictability of the true criterion values ​​from the indications and R S the consistency with which a judge makes the Using cues for the judgments allows this method to determine measures of these psychologically significant variables. The measures r a , R E and R S are calculated directly by means of the regression analyzes; C is determined in a second step (if the regression residuals have been determined). By inserting it into the lens model equation, the matching index G can then be calculated in the third step .

Critical appreciation and further development

The methods developed from the conception of the lens model have had a decisive influence on judgment research and, especially since the 1950s, have spurred the debate between critics and proponents of "clinical" or "intuitive" expert judgments versus statistical judgment models. A great advantage of the lens model is its flexibility, which enables the concept and the method to be applied to almost any judgment context. Critically on the conceptual level, it was noted that the regression equation on the side of the judge characterizes the judgment process by determining the importance of individual indications for the judgments, but that it does not provide any information about the cognitive sub-processes of judgment. Methodologically, it has been criticized that the coefficients of the use of cues can only be interpreted relative to the examined sample of objects and therefore do not represent an absolute measure of the importance of these cues for the judgments. So-called Neo-Brunswikians have retained the concept of judgment formation on the basis of cues, but go beyond traditional analysis and formulate more explicit cognitive process models of cue processing.

Individual evidence

  1. ↑ Karelaia, N., & Hogarth, RM (2008). Determinants of linear judgment: A meta-analysis of lens model studies. Psychological Bulletin, 134 (3), 404-426.
  2. ^ Brehmer, A., & Brehmer, B. (1988). What have we learned about human judgment from thirty years of policy capturing? In B. Brehmer & CRB Joyce (Eds.), Human judgment: The SJT view. (Pp. 75-114). Oxford, England: North Holland.
  3. ^ Brunswik , E. (1944). Distal focussing of perception: Size-constancy in a representative sample of situations. Psychological Monographs, 56 (1), i-49.
  4. ^ Brunswik , E. (1952). The conceptual framework of psychology. (Int. Encycl. Unified Sci., V. 1, no. 10.). Oxford, England: Univ. Chicago Press.
  5. ↑ Hammond, KR (1955). Probabilistic functioning and the clinical method. Psychological Review, 62 , 255-262.
  6. ^ Brehmer, B., & Joyce, CRB (Eds.) (1988). Human judgment: The SJT view. Oxford, England: North Holland.
  7. ^ Tucker, LR (1964). A suggested alternative formulation in the developments by Hursch, Hammond, and Hursch, and by Hammond, Hursch, and Todd. Psychological Review, 71 (6), 528-530.
  8. ^ Hoffman, PJ (1960). The paramorphic representation of clinical judgment. Psychological Bulletin, 57 (2), 116-131.
  9. ↑ Bröder, A. (2000). A methodological comment on behavioral decision research. Psychological Contributions, 42 (4), 645-662.
  10. ↑ Fiedler , K. (1996). Explaining and simulating judgment biases as an aggregation phenomenon in probabilistic, multiple-cue environments. Psychological Review, 103 (1), 193-214.
  11. ↑ Gigerenzer , G., Hoffrage, U., & Kleinbölting, H. (1991). Probabilistic mental models: A Brunswikian theory of confidence. Psychological Review, 98 (4), 506-528.