Is there a Sunk Cost Effect in Committed Relationships?

The sunk cost effect occurs when a prior investment in one option leads to a continuous investment in that option, despite not being the best decision. The aim of the present paper was to study the role of the sunk cost effect in committed relationships. In Experiment 1, participants (N = 902) were presented with an unhappy relationship scenario in which they needed to make a choice: to stay or end the relationship. Results showed that the likelihood of participants staying in the relationship was higher when money and effort, but not time, had been previously invested in that relationship. In Experiment 2, the time investment was manipulated and the sunk cost was evaluated using a different methodology. Specifically, instead of having a dichotomous decision, participants (N = 275) choose how much time they would be willing to invest in the relationship. Results revealed a sunk time effect, that is, participants were willing to invest more time in a relationship in which more time had already been invested.     

The sunk cost effect across species: A review of persistence in a course of action due to prior investment

The sunk cost effect is the bias or tendency to persist in a course of action due to prior investments of effort, money or time. At the time of the only review on the sunk cost effect across species (Arkes & Ayton, 1999), research with nonhuman animals had been ecological in its nature, and the findings about the effect of past investments on current choice were inconclusive. However, in the last decade a new line of experimental laboratory‐based research has emerged with the promise of revolutionizing the way we approach the study of the sunk cost effect in nonhumans. In the present review we challenge Arkes and Ayton's conclusion that the sunk cost effect is exclusive to humans, and describe evidence for the sunk cost effect in nonhuman animals. By doing so, we also challenge the current explanations for the sunk cost effect in humans, as they are not applicable to nonhumans. We argue that a unified theory is called for, because different independent variables, in particular, investment amount, have the same influence on the sunk cost effect across species. Finally, we suggest possible psychological mechanisms shared across different species, contrast and depreciation, that could explain the sunk cost effect.     

Do birds of a feather flock together? A test of the similarities within and the differences between five occupations

Two sets of samples of 40 women from each of five different occupations were subjected to J. H. Ward's (Journal of the American Statistical Association, 1963, 58, 236–244) hierarchical cluster analysis, using an index of similarity from Kuder, 1977, Kuder, 1980, 40, 1–8). The five-group solution in neither case adequately recaptured the occupational membership of the original groups. This finding was interpreted to mean that there may be important differences in the interests of person in the same occupation, which in turn has implications for occupational scaling on interest inventories and theories of career development and counseling.     

Effects of schedule and magnitude of reinforcement under conditions of thirst motivation

In contrast to a recent finding (Macdonald, G. E., & De Toledo, L. Learning and Motivation, 1974, 5, 288–298.) the results of three experiments investigating various partial reinforcement (PRF) manipulations under conditions of thirst motivation demonstrated strong similarity to analogous manipulations involving food reward. Specifically, for animals receiving water reinforcement, PRF was shown to generate greater resistance to extinction than continuous reinforcement (Expt 1 & 3), the schedule of reinforcement was shown to interact with level of acquisition (Expt 1 and 2), and the magnitude of the partial reinforcement extinction effect was shown to be a function of reward magnitude (Expt 3). These results provide strong evidence that mechanisms which operate in partial reinforcement situations are highly similar, regardless of the type of appetitive reinforcement.     

Inaction inertia, the sunk cost effect, and handedness: Avoiding the losses of past decisions

Strength of handedness, or the degree to which an individual prefers to use a single hand to perform various tasks, is a neurological marker for brain organization and has been shown to be linked to episodic memory, attribute framing, and anchoring, as well as other domains and tasks. The present work explores the relationship of handedness to both inaction inertia (the inclination to resist an action after previously bypassing a similar action) as well as to the sunk cost effect (the tendency to continue to engage in a behavior after an initial investment of time or money has been made). In Experiment 1, mixed-handers displayed a larger inaction inertia effect than strong-handers. In Experiment 2, they displayed a larger sunk cost effect than strong-handers. Experiments 3 and 4 extended the sunk cost finding into a different domain and explored how mixed- and strong-handers react to additional information designed to increase the comparative advantage of terminating, rather than continuing, a failed project. Overall, we found that mixed-handers were more likely to show inertia effects because of an increased aversion to losses. The results of Experiment 4 suggest that, when provided with additional information that made it clear that continuing a project would be a greater loss than terminating it, mixed-handers no longer showed a larger sunk cost effect than strong-handers, highlighting the importance of carefully considering exactly how sunk cost scenarios are worded and providing additional information on how mixed- and strong-handers differ in belief updating.     

Sunk cost of local elected representatives in situation of fiscal competition: An example of escalating commitment

Introduction

Competition between French local communities to build infrastructures to attract enterprises and residents could sometimes result in irrational public spending.

Developmental aspects of the generation effect

Children from Grades 2, 3, 5, and 7 (7.6, 9, 11, and 13 years of age, respectively) were required to generate or study exemplars of semantic categories (semantic orientation) or rhymes to stimulus words (phonetic orientation). Each child then participated in one of three retention tests: free recall, standard recognition, and rhyme recognition. The results indicated that the developmental emergence of the “generation effect” (C. E. McFarland, Jr., T. J. Frey, & D. D. Rhodes, Journal of Verbal Learning and Verbal Behavior, 1980, 19, 210–225; N. J. Slamecka & P. Graf, Journal of Experimental Psychology: Human Learning and Memory, 1978, 4, 592–604) was dependent on both encoding orientation and the type of retention task employed. A substantial generation effect first emerged (7-year-olds) for standard recognition in the semantic condition. A similar effect for recall was evident for 9-year olds, but not for younger children. Internal stimulus generation became a strong memory facilitator for phonetically encoded items at age 11 for standard recognition, but not until age 13 for recall. The results of the rhyme recognition test indicated that internal generation facilitated “transfer-appropriate processing” (C. D. Morris, J. D. Bransford, & J. J. Franks, Journal of Verbal Learning and Verbal Behavior, 1977, 16, 519–533) only for seventh graders (age 13). The processes underlying this developmental pattern were discussed.     

Measurement foundations for multiattribute psychophysical theories based on first order polynomials

The class of first order polynomial measurement representations is defined, and a method for proving the existence of such representations is described. The method is used to prove the existence of first order polynomial generalizations of expected utility theory, difference measurement, and additive conjoint measurement. It is then argued that first order polynomial representations provide a deep and far reaching characterization of psychological invariance for subjective magnitudes of multiattributed stimuli. To substantiate this point, two applications of first order polynomial representation theory to the foundations of psychophysics are described. First, Relation theory, a theory of subjective magnitude proposed by Shepard (Journal of Mathematical Psychology, 1981, 24, 21–57) and Krantz (Journal of Mathematical Psychology, 1972, 9, 168–199), is generalized to a theory of magnitude for multiattributed stimuli. The generalization is based on a postulate of context invariance for the constituent uniattribute magnitudes of a multiattribute magnitude. Second, the power law for subjective magnitude is generalized to a multiattribute version of the power law. Finally, it is argued that a common logical pattern underlies multiattribute generalizations of psychological theories to first order polynomial representations. This abstract pattern suggests a strategy for theory construction in multiattribute psychophysics.     

Informational cues and attributions based on role behavior

A recent experiment by Messick and Reeder (Journal of Experimental Social Psychology, 1972, 18, 482–491) attempted to extend Jones, Davis, and Gergen's (Journal of Abnormal and Social Psychology, 1961, 63, 302–310) classic finding that out-of-role behavior.is more informative for person perception than in-role behavior. It is argued, however, that this study confounded two variables, role performance and occupation. Evidence is presented that the occupation variable alone could have produced Messick and Reeder's results. Both variables seem to affect attributions. The importance of these findings for relating attribution theory and role theory is discussed.     

Perceived infant temperament,locus of control,and maternal physiological response to infant gaze

Maternal physiologic response was used as an index of maternal attention to infant eye contact. Thirty-six women, each of whom had a 3-month-old infant, were shown 10-sec silent images of a 3-month-old infant on a videotape monitor. In one condition the en face position offered eye contact, in the other condition eye contact was not possible. Each woman viewed a sequence of six identical episodes of one condition followed by six episodes of the other condition. Heart rate and skin conductance were recorded continuously during the session. Skin conductance response did not differentiate between reciprocated and unreciprocated gaze. Cardiac response elicited by reciprocated gaze viewed first shifted from a predominately orienting response on early trials to an acceleratory response on Trial 6. The acceleratory response was most pronounced in those women identified as Externals by the Locus of Control Inventory (Rotter, J. B. Psychological Monographs, 1966, 80, [No. 609]). Initial viewing of the infant looking away elicited a predominantly deceleratory heart-rate response that did not habituate. Mothers who described their infants as being difficult as assessed by the questionnaire method (Carey, W. B. Journal of Pediatrics, 1970, 77, 188–194) were physiologically less sensitive to the change from the infant's averted gaze to its direct gaze. Results are discussed in terms of the role of direct and averted gaze during social transactions.     

Sequencing of vocational development stages: Further studies

This paper reports two studies which support the prediction derived from Hershenson's (Journal of Counseling Psychology, 1968, 15, 23–30) life-stage vocational development model that average scores on Self-differentiation (worker self-concept and motivation) would exceed those on Competence (work habits, skills, and interpersonal relations), which in turn would exceed those on Independence (appropriateness and crystallization of vocational goals). The first study involved ratings by project staff on an inner city, socially disadvantaged population, and the second study involved self-ratings by individuals who had changed occupations in midcareer. Findings are consistent with those reported by Hershenson and Langbauer (Journal of Counseling Psychology, 1973, 20, 519–521) on a population of deaf clients.     

The Sunk Costs Fallacy or Argument from Waste

This project tackles the problem of analyzing a specific form of reasoning called sunk costs in economics and argument from waste in argumentation theory. The project is to build a normative structure representing the form of the argument, and then to apply this normative structure to actual cases in which the sunk costs argument has been used. The method is partly structural and partly empirical. The empirical part is carried out through the analysis of case studies of the sunk costs argument found in business decision-making, as well as other areas like medical decision-making and everyday conversational argumentation. The structural part is carried out by using existing methods and techniques from argumentation theory, like argumentation schemes. The project has three especially significant findings. First, the sunk costs argument is not always fallacious, and in many cases it can be seen to be a rational precommitment strategy. Second, a formal model of argumentation, called practical reasoning, can be constructed that helps a rational critic to judge which sunk costs arguments are fallacious and which are not. Third, this formal model represents an alternative model of rationality to the cost-benefit model based on Bayesian calculation of probabilities. This alternative model is called the argumentation model, and it is based on interpersonal reasoning in dialogue as the model of rational thinking. This model in turn is based on the underlying notion of commitment in dialogue.     

The halo effect revisited: Forewarned is not forearmed

The purpose of the present research was to determine if forewarning subjects about the halo effect eliminated the effect or made people aware of its impact. The research was a replication and extension of R. E. Nisbett and T. D. Wilson's (Journal of Personality and Social Psychology, 1977, 35, 250–256) experiment on the halo effect. Subjects viewed an interview of a college instructor who was either warm or cold and then rated his physical appearance, mannerisms, and French accent. Some subjects were told to introspect about their cognitive processes while viewing the interview and were told that the purpose of the study was to see how aware they were of the determinants of their ratings. Other subjects were also told specifically what the halo effect was and given a motivation either to show it or not to show it. In addition, a measurement technique different from that used by Nisbett and Wilson was utilized to allow a more precise measurement of awareness, and a new cover story was used to make the task more involving and important to subjects. Despite these attempts to eliminate the halo effect (or, at a minimum, to make people aware of it), the results indicated that subjects in all conditions were very susceptible to it. That is, subjects who viewed the warm version of the interview rated the instructor's appearance, mannerisms, and accent significantly higher than did those who viewed the cold version, even when informed and forewarned about this effect. The forewarning and introspection instructions also had no impact on subject's awareness of the halo effect. Subjects in all conditions indicated that their liking for the instructor had had only a minimal effect on their ratings.     

On mediation

The mediation-nonmediation dichotomy proposed by the Kendlers (Psychological Review, 1962, 69, 1–16) and subscribed to by Cole and Medin (Journal of Experimental Child Psychology, 1973, 15, 352–355) is reviewed and critiqued. It is proposed that the appropriate question for research is how children mediate, not when they mediate.     

Sunk Cost and Commitment to Medical Treatment

The influence of prior, irretrievable, investment (sunk cost) on commitment to medical treatment was investigated. Three studies were run investigating the influence of sunk cost in the form of money, time, and effort. A total of 637 participants (314 male) with a mean age of 19.58 years were recruited from an undergraduate population. A computer program simulated the process of arranging a course of physiotherapy. Participants invested one of three amounts of sunk cost (under budget, on budget, or over budget) into arranging sessions with a chiropractor. Participants then decided how much time they wished to commit to these chiropractor sessions or to an alternative treatment with a better chance of success. Results revealed a significant effect of invested money, a significant effect of invested effort, but no effect of invested time. Invested money produced a sunk cost effect, while invested effort appeared to exert influence via cognitive dissonance. The implications for healthcare decision-making are discussed.     

The role of probability of success estimates in the sunk cost effect

The sunk cost effect is manifested in a tendency to continue an endeavor once an investment has been made. Arkes and Blumer (1985) showed that a sunk cost increases one's estimated probability that the endeavor will succeed [p(s)]. Is this p(s) increase a cause of the sunk cost effect, a consequence of the effect, or both? In Experiment 1 participants read a scenario in which a sunk cost was or was not present. Half of each group read what the precise p(s) of the project would be, thereby discouraging p(s) inflation. Nevertheless these participants manifested the sunk cost effect, suggesting p(s) inflation is not necessary for the effect to occur. In Experiment 2 participants gave p(s) estimates before or after the investment decision. The latter group manifested higher p(s), suggesting that the inflated estimate is a consequence of the decision to invest. Copyright © 2000 John Wiley & Sons, Ltd.     

沉没成本效应的产生根源与影响因素

沉没成本效应是指决策者的决策行为因受沉没成本影响而产生的一种非理性决策现象。针对这一决策偏差的产生根源,研究者从认知、动机和神经三个角度提出了解释。沉没成本效应的影响因素包括沉没成本特性、情境因素、个体差异和文化差异等。基于先前研究存在的问题和不足,沉没成本效应的未来研究应着眼于改进研究方法,探究产生根源,关注行为沉没成本和加强应用研究。     

Escalation and De-escalation of Commitment in Response to Sunk Costs: The Role of Budgeting in Mental Accounting

Research has traditionally assumed that people increase investment (or "escalate commitment") in response to previous investments (sunk costs). This paper presents several demonstrations which show that people will incorrectly de-escalate investment in response to sunk costs. I propose that people set mental budgets to control their resource expenditures: they set a budget for a class of expenses and track their investments against their budget. A lab study with real monetary incentives shows support for de-escalation and supports a specific rule for how people set budgets - based on the breakeven of total costs and total benefits. The budgeting process suggests that people are only likely to escalate commitment when they fail to set a budget or when expenses are difficult to track. The later part of the paper organizes the previous literature on escalation around these processes and provides additional experiments to illustrate each point. For example, I argue that previous demonstrations that have shown errors of escalation exclusively involve "incidental" investments that are difficult to track. A study in the current paper shows that people are more willing to invest time than money to salvage a monetary sunk cost and more willing to invest money than time to salvage a sunk cost of time, even when the time and money investments are of equal value. The paper concludes by discussing the rationality of escalation and de-escalation.     

Values,preferences, needs,work ethics,and orientations to work: Toward a conceptual and empirical integration

There have been comparatively few attempts on either the conceptual or empirical level to integrate values, preferences, needs, work ethics, and orientations to work. A rationale outlining the way these concepts can be related to each other, is presented. It is maintained that these concepts are categories of relations between humans and work. These categories are qualitatively similar and differ principally in the level of generality of the characteristics of work considered. Complete-link cluster analysis is suggested as a method of analysis compatible with the proffered rationale. Two studies are presented with data gathered during the development of two versions of Work Aspect Preference Scale (WAPS). The samples for the two studies both comprised senior high school students, controlled for sex and age, who were administered the WAPS. Study 1 had a sample of 532 while Study 2 had a sample size of 482. Both sets of data were cluster analyzed. Results indicated a very substantial amount of agreement between the two samples, particularly at the more general levels of clustering. The bipartite and tripartite groupings of vocational needs suggested by L. H. Lofquist and R. V. Dawis (Journal of Vocational Behavior, 1978, 12, 12–19) were confirmed as were the findings of J. Mortimer (Sociology of Work and Occupations, 1975, 2, 29–53). Other clusters were also identified with findings from previous research. It is concluded that the investigation of preferences at different levels of generality and the interrelation between these levels shows great promise for both the researcher and the counselor and therefore merits more research effort.     

Further evidence for meaning change in impression formation

Hamilton and Zanna (Journal of Personality and Social Psychology, 1974, 29, 649–654) demonstrated that the meaning of an attribute as rated on connotatively related scales changed as a function of context desirability. This finding was viewed as supporting a meaning change interpretation of context effects in impression formation. Kaplan (Memory and Cognition, 1975, 3, 375–380, Exp. 1) found that similar changes also occurred on scales unrelated in meaning to the test attribute and argued that changes on both kinds of scales were due to halo effects produced by the contexts. Controlling for possible scaling artifacts, the first experiment reported here showed that substantially greater changes occurred on related than on unrelated scales. Kaplan (1975, Exp. 2) also found that the magnitude of context-induced differences in judgments of trait likability was not increased by creating contexts denotatively related to the test attributes. Adding measures of connotative meaning, the second experiment in this paper showed that substantially greater context-induced changes did occur on related meaning scales and, in contrast to Kaplan, on trait likability for the denotatively related contexts. The results of each experiment are consistent with a meaning change position but would not be expected on the basis of the halo effects model.