Tipultech logo

Illusory correlations

Author: Dr Simon Moss


Often, individuals hear more information or descriptions about one company or group than another. They might, for example, gather more information about Microsoft than Apple. Suppose that 75% or more of the information or descriptions of both companies or groups is positive.

In this instance, individuals will tend to prefer the company or group in which they received most of the information. They will prefer Microsoft to Apple in this example, called a distinctiveness-based illusory correlation (e.g., Johnson, Mullen, Carlson, & Southwick, 2001), and sometimes referred to as covariation bias in the literature on phobias and anxiety (e.g., De Jong, 1993).

According to Johnson, Mullen, Carlson, and Southwick (2001), individuals tend to focus their attention on information that is not common. In this example, individuals will tend to direct their attention to Apple and to negative information. Because of this bias in their attention, they will become more aware of the shortfalls and deficiencies of the company or group in which less information is presented. Other scholars offer a different explanation, however (Fiedler, 2008).

Examples of illusory correlations

Distinctiveness based illusory correlations

In a typical demonstration, participants receive two facts about a set of individuals. One fact is often the group to which they belong, such as male or female. The second fact describes some behavior, such as whether or not they recycle.

One group is always more frequent than is the other group. Furthermore, one behavior is more frequent than is another behavior. Participants tend to assume the larger group is more inclined to engage in the more frequent behavior (e.g., Acorn, Hamilton, & Sherman, 1988& Hamilton & Gifford, 1976& Fiedler, 1991& Johnson & Mullen, 1994& McConnell, Sherman, & Hamilton, 1994& Stroessner, Hamilton, & Mackie, 1992).

To illustrate, suppose information about a set of males and females is presented. Furthermore, assume this set comprises more males than females. In addition, suppose that individuals in this set are more likely to recycle than not recycle. Participants will usually assume that males are more likely than females to recycle.


Illusory correlations might underpin, amplify, or reinforce some phobias. Tomarken, Mineka, and Cook (1989), for example, developed a paradigm to explore this possibility. In this paradigm, various pictures are presented, some of which correspond to the phobia of participants, such as spiders to individuals with arachnophobia. Furthermore, each picture is paired with a positive or negative event, such as mild electric shocks. Individuals who exhibit phobias tend to overrate the frequency with which the phobic stimuli were paired with negative outcomes, called a covariation bias (see also Pauli, Montoya, & Martz, 1996).

Similarly, Hermann, Ofer, and Flor (2004) showed that social phobia might also generate illusory correlations. In this study, descriptions of ambiguous social events, threatening animals, or nature scenes were presented. These descriptions randomly coincided with positive, negative, or neutral facial expressions. Participants with social phobia overestimated the frequency with which social cues coincided with negative facial expression. Conceivably, these individuals are more sensitive to this association between social cues and negative outcomes--perhaps because, in the past, these social environments often culminated in unfavorable experiences.

One-shot illusory correlations

Risen, Gilovich, and Dunning (2007) showed that illusory correlations can even be derived from a single event, called one-shot illusory correlations. In particular, they showed that one unusual event perpetuated by a member of an unfamiliar group is sufficient to generate these correlations.

Risen, Gilovich, and Dunning (2007) conducted a series of studies to examine these illusory correlations. In one study, participants read a series of sentences, such as "Ben, a Jehovah's Witness, owns a pet sloth". The time that individuals devoted to reading each sentence was assessed. Participants dedicated more time to reading uncommon rather than common behaviors--but only if the protagonist was a member of an unfamiliar group.

That is, according to this study, only the combination of an unusual behavior and an unfamiliar group attracted significant attention. Hence, these combinations are especially salient. Consistent with this proposition, a subsequent study showed that individuals are more likely to recall unusual behaviors that were perpetrated by unfamiliar rather than family groups (Risen, Gilovich, & Dunning, 2007). The final study showed these illusory correlations do indeed shape subsequent attitudes towards the unfamiliar group, as gauged by implicit measures.

Other illusory correlations

Many researchers, such as Whitson and Galinsky (2008), have uncovered other inclinations in individuals that resemble illusory correlations. According to Whitson and Galinsky (2008), superstition reflects illusory correlations. In particular, when individuals experience a limited sense of control, they experience the need to uncover patterns, as a means to derive meaning and certainty. Hence, they observe correlations between their behaviors, such as averting ladders, and outcomes, which reinforce their superstitions.

Furthermore, individuals also seem to observe patterns in random scenes. In one study, for example, a series of photographs were presented, in which the scene is grainy, and could hardly be discerned. Indeed, in some of the pictures, all images were removed& the picture, in essence, was a random array of grey blotches. Nevertheless, most participants claimed to discern a pattern in the picture--especially if they experienced a limited sense of control (Whitson & Galinsky, 2008).

Overestimation of sizeable, expensive, or demanding initiatives

Often, individuals need to decide which of two or more initiatives should be introduced to resolve some problem. For example, they might need to choose between two or more exercise routines or between two or more dieting regimes. Ideally, if all these initiatives generate the same effects, they should choose the alternative that is the least demanding. They should, for example, prefer the exercise routine that is not arduous or the dieting regime that is not too burdensome. However, partly because of illusory correlations, individuals are often more compelled by the most expensive, demanding, or sizeable initiatives (Fiedler, Freytag, & Unkelbach, 2011).

To illustrate, in one study, conducted by Fiedler, Freytag, and Unkelbach (2011), participants received information about the effects of some dieting program. Some participants were told the dieting program was very effective, increasing some index of health from about 2.5 to 3.6 units on average in a sample of 50 or 150 people. Other participants were told the dieting program was mildly effective, increasing this index of health from about 2.5 to 3.0 units.

Furthermore, the magnitude of this diet was also manipulated. Some participants were informed that only the 15% of people who adhered to this diet the most rigorously were examined& hence, the study, in essence, examined only an arduous diet. Other participants were informed the entire sample, including some people who did not adhere to this diet, were examined as well& hence, in this instance, the study essentially examined a modest diet. Finally, participants were asked to indicate the extent to which they felt the diet was effective. That is, on a rating scale, they specified the degree to which the effect of this diet was strong as well as the extent to which the study was convincing.

Obviously, participants should be more compelled by diets that enhanced health appreciably rather than marginally. In addition, and perhaps more subtly, participants should be more compelled by the study if the diet was modest rather than arduous. A diet that is successful, even if not followed entirely, must be especially effective.

However, participants were actually more compelled by studies that examined the arduous diet, instead of the modest diet, even if these two alternatives demonstrated the same effect. Presumably, consistent with the rationale of illusory correlations, the arduous diet is more salient and, thus, assumed to be highly correlated with the improvement in health.

Other biases might also translate into flawed judgments. For example, people are also more compelled when the magnitude of some treatment matches the magnitude of some effect. They will be persuaded, for example, by arduous diets that improve health considerably but also persuaded by modest diets that improve health marginally (Fiedler, Freytag, & Unkelbach, 2011).

Indeed, Fiedler, Freytag, and Unkelbach (2011) found some evidence of this possibility. Participants read about either about an extensive or short psychotherapy with either modest or sizeable benefits. Some participants were more compelled by these treatments if the magnitude of the therapy matched the magnitude of the effect--such as a short therapy with a modest effect or a long therapy with a sizeable effect.

Mechanisms that underpin illusory correlations

Some researchers assume that memory of infrequent events, such as females who do not recycle, are more salient, underpinning this illusory correlation (Johnson, Mullen, Carlson, & Southwick, 2001). Nevertheless, studies that assess the memory of participants seem to challenge this assumption: Participants do not seem to remember the group to which individuals who engaged in a specific behavior belonged (Meiser, 2003& see also Meiser & Hewstone, 2001, 2006).

Instead, participants seem to derive these correlations from the base or marginal rate of each category (Fiedler, 2008). To illustrate, consider the example in which a sample comprises more men than women and more individuals who recycle than waste resources. In this example, participants might perceive males as more common than females and recycling behavior as more common than wasteful behavior. From this information alone, they might conclude that males are more inclined to recycle. Consistent with this premise, participants have been shown to derive these conclusions when only the base rates, rather than information about each individual, is presented (e.g., McGarty, Haslam, Turner, & Oakes, 1993).


A variety of factors, such as the extent to which individuals experience a sense of control, can affect the magnitude of illusory correlations. For example, as Fiedler, Freytag, and Meiser (2009) maintained, these illusory correlations are especially pronounced when participants are encouraged to orient their attention towards the base rates--towards the observation that one group and behavior is especially common, for example. Second, these illusory correlations may be amplified when the two facts about each person are presented in sequence rather than in parallel (e.g., see Fiedler & Freytag, 2004)--to override the attention to genuine contingencies (Fiedler, Freytag, & Meiser, 2009).

Processing effort

Several studies indicate that any factors that curb effort and deliberation tend to amplify illusory correlations. That is, when deliberation is curbed, individuals invoke economical heuristics or algorithms, which have been shown to exacerbate this illusion. They might, for example, only consider the base rates. Gordon (1997), for example, showed that illusory correlations are more pronounced when individuals do not feel alert. Specifically, these illusory correlations are amplified when individuals who are usually more alert in the morning operate at night--and vice versa.

Similarly, Gordon (1997) demonstrated that illusory correlations are more pronounced in individuals who exhibit a high need for personal structure. These individuals seek clarity rather than ambiguity and often utilize heuristics and algorithms to form rapid judgments.

Sense of control

As Whitson and Galinsky (2008) show, when individuals do not feel a sense of control, they are more likely to detect illusory correlations and patterns. To reduce any sense of control, in one study, participants received random feedback that was actually unrelated to their actual performance. In another study, participants were instructed to recall a time in which they felt no sense of control.

In a series of studies, when participants did not feel a sense of control, they were more likely to recognize illusory correlations. Furthermore, self affirmation, in which participants reflected upon their key values, mitigated this effect of limited control.

According to Whitson and Galinsky (2008), when individuals do not experience a sense of control, they feel the need to seek meaning and purpose. They attempt to uncover patterns and regularities to instil a sense that life is coherent and meaningful.


These illusory correlations are embedded within a broader framework called pseudocontingencies. One of most important facets of learning is the capacity to distill correlations between events (Fiedler, Freytag, & Meiser, 2009). To refine their social skills, individuals need to ascertain the correlation between the words they express and the subsequent reactions of the listener. To improve their health, they need to determine the correlation between the foods they consume and their subsequent bodily sensations. Such information enables individuals to ascertain which actions they should select in a specific circumstance and to predict future events from the immediate environment. Indeed, the capacity to identify these correlations is considered a key facet of inductive intelligence (e.g., Allan, Hannah, Crump, & Siegel, 2008& Arieh & Algom, 2002).

Nevertheless, as Fiedler, Freytag, and Meiser (2009) emphasized, these correlations are difficult to distill. To illustrate, suppose individuals wanted to ascertain the correlation between the foods they consume and their subsequent bodily sensations. To estimate this correlation, individuals would like to be able to compile a protracted list of specific ingredients they ingested and the bodily sensation that followed each ingredient. They cannot compile this list, however. They typically consume many ingredients, and indeed several dishes, within a limited timeframe. They also cannot even ascertain whether specific bodily sensations, such as a stomach ache at night, can be ascribed to dinner, lunch, or breakfast.

Individuals, however, might extract this correlation between the food they consume and the sensations they experience from other information. They might, for example, recognize they consume more chocolate than any other ingredient. They might also realize they experience stomach aches more than any other bodily reaction. As a consequence, they tend to infer that chocolate might elicit these stomach aches. This correlation has not been derived from specific pairings of food and reactions--but from other properties of the data. Correlations derived from other properties of the data are called pseudocontingencies (Fiedler, 2008& Fiedler & Freytag, 2004& Fiedler, Freytag, & Meiser, 2009).

Aggregation bias

To appreciate pseudocontingencies, the concept of aggregation bias needs to be appreciated. Specifically, the correlation between two variables often depends on whether the researcher is investigating individuals, teams, organizations, regions, societies, or nations (Hammond, 1973). This principle is often called the aggregation bias.

A seminal example of this aggregation bias was discovered by Robinson (1950). To illustrate this discovery, suppose 100 individuals are subjected to an IQ test. Research usually shows the correlation between their skin color and intelligence is low& skin color does not seem to correlate appreciably with intelligence at the level of individuals (Robinson, 1950).

In contrast, suppose instead 100 districts in the United States was assessed. In this instance, the percentage of individuals with black skin in these districts is inversely, and appreciably, correlated with the average level of intelligence in these regions. Districts with many African Americans tend to generate lower levels of intelligence on average. Skin color does seem to correlate markedly with intelligence at the level of districts ((Robinson, 1950).

This aggregation bias arises because the factors that determine these correlations varies across the levels of analysis. At the individual level, genetics and family background, perhaps, largely affect the correlation between skin color and intelligence. At the district level, other factors, such as funding, might primarily influence the correlation between skin color and intelligence.

In short, the correlation between skin color and intelligence at the individual level, sometimes called the total correlation, depends on two factors. The first factor is the correlation between skin color and intelligence at the district or group level, often called ecological correlation. The second factor is the average correlation between skin color and intelligence at the individual level within each district or group, called the partial correlation.

The derivation of correlations from ethological correlations

Several studies have demonstrated that individuals derive pseudocontingencies. In particular, some studies have shown the total correlation between two variables, such as skin color and IQ at the individual level, is partly derived from ethological or aggregated correlations.

One paradigm has been used convincingly to demonstrate this proposition. Participants assume the role of a teacher, attempting to ascertain whether attitudes towards two issues, such as freedom of science and benefits of humanism, are correlated across students (Fiedler, Freytag, & Unkelbach, 2007& Fiedler, Walther, Freytag, & Plessner, 2002). The information is biased to ensure no total correlation at the individual level. However, the students are divided into two categories& at this aggregated level, a correlation is introduced. If this correlation is positive, participants often assume the correlation is also positive at the individual level.

The derivation of correlations from single ecologies

As Fiedler, Freytag, and Meiser (2009) highlight, pseudocontingencies can be derived from instances in which the individuals or observations correspond to a single ecology or group. To illustrate, a teacher might be informed that 75% of the students in a class are boys and 75% of the students prefer Mathematics to English. That is, the base rate of both boys and preferences towards Mathematics is high. The teacher, therefore, will often assume that boys prefer Mathematics to English. Indeed, many studies have shown that base rates determine the perceived correlations (e.g., Allan, 1993& DeHouwer & Beckers, 2002& Hamilton, 1981& White, 1995).

Related findings or heuristics

Consequence-cause matching

As Leboeuf and Nortonb (2012) showed, individuals show an interesting tendency called consequence-cause matching. In particular, if some event translates into a severe consequence, people assume the cause of this event must be enormous as well.

To illustrate, in one study, participants were told about a student who was writing an assignment for his university. The computer crashed and the student lost the assignment. In one condition, participants were told the consequences of this event were severe: The student failed the course and could not secure a job afterwards. In the other condition, participants were told the consequences of this event were modest: The professor granted the student an extension. Finally, participants needed to decide whether a widespread virus, a severe cause, or merely a malfunctioning fan, a modest cause, was the source of the crash.

Interestingly, if the consequences were severe, and the student failed the course, participants tended to assume the widespread virus caused the crash. If the consequences were modest, and the student was granted an extension, participants were more likely to assume the malfunctioning fan caused the crash. Yet, of course, the impact of this crash was independent of the cause of this crash--but depended only on the decision of this professor.

A subsequent study was the same, except participants were told the student had used antivirus software. If the consequences of this crash were severe, participants were more likely to rate the antivirus software unfavorably--even though this software could not have affected the consequences of this crash.

Similarly, in another study, participants were told that a president of a small country was assassinated--a person who had been criticized by a British newspaper. This event sparked a war with Britain that was either prolonged, killing many people, or quelled rapidly by Britain, killing few people. If the war killed many people, participants were more likely to assume the president was killed by a large conspiracy. If the war killed few people, participants were more likely to assume the president was killed by a lone gunman.

Likewise, in an unusual study, participants were informed that a disease had spread to all the animals of a zoo. Some participants were then told many animals died. Other participants were told the caretakers were able to intervene and diminish the effect of this disease. If the consequences were severe and many animals died, participants felt the disease probably emanated from a large animal, such as a bear& otherwise, participants felt the disease probably emanated from a small animal, such as a rabbit.

In addition to matching the magnitude of consequences with causes, participants like to match the valence as well. To demonstrate, in one story, participants heard about a conflict between a married couple. The man felt remorseful as he left the house that morning and therefore bought flowers for his wife, arriving late to work. Some participants were told the man was dismissed for his tardiness. Other participants were told that his meeting was postponed& this delay enabled the man to improve his preparation for the meeting, ultimately translating to a promotion. In general, participants ascribed the negative event, the dismissal, to the conflict. In contrast, they ascribed the positive event, the promotion, to his purchase of flowers.

According to Leboeuf and Nortonb (2012), consequence-cause matching can be ascribed to a simple argument. Specifically, people usually likely to perceive the world as predictable and coherent. Because the magnitude and valence of consequences and causes tend to match, and therefore this association is often salient, this heuristic can be applied to understand the source of events. Consequently, the world seems more predictable and coherent.

Consistent with this premise, when people are not as motivated to perceive the world as predictable or coherent, consequence-cause matching dissipates. That is, in one study, if people are informed the world is predictable, and hence this need is fulfilled, consequence-cause matching was not as pronounced.


Acorn, D. A, Hamilton, D. L., & Sherman, S. J. (1988). Generalization of biased perceptions of groups based on illusory correlations. Social Cognition, 6, 345-372.

Allan, L. G. (1993). Human contingency judgments: Rule based or associative? Psychological Bulletin, 114, 435-448.

Allan, L. G., Hannah, S. D., Crump, M. J. C., & Siegel, S. (2008). The psychophysics of contingency assessment. Journal of Experimental Psychology: General, 137, 226-243.

Amin, J. M., & Lovibond, P. F. (1997). Dissociations between covariation bias and expectancy bias for fear-relevant stimuli. Cognition and Emotion, 11, 273-289.

Arieh, Y., & Algom, D. (2002). Processing picture-word stimuli: The contingent nature of picture and of word superiority. Journal of Experimental Psychology: Learning, Memory, and Cognition, 28, 221-232.

DeHouwer, J., & Beckers, T. (2002). A review of recent developments in research and theories on human contingency learning. Quarterly Journal of Experimental Psychology: Comparative and Physiological Psychology, 55(B), 289-310.

De Jong, P. J. (1993). Covariation bias in phobia: Mere resistance to preexperimental expectancies? Behavior Therapy, 24, 447-454.

De Jong, P. J., Merckelbach, H., & Arntz, A. (1990). Illusory correlation, on-line probability estimates, and electrodermal responding in a (quasi)-conditioning paradigm. Biological Psychology, 31, 201-212.

De Jong, P. J., Merckelbach, H., B?gels, S., & Kindt, M. (1998). Illusory correlation and social anxiety. Behaviour Research & Therapy, 36, 1063-1073.

Fiedler, K. (1986). Person memory and person judgments based on categorically organized information. Acta Psychologica, 61, 117-135.

Fiedler, K. (1991). The tricky nature of skewed frequency tables: An information loss account of "distinctiveness-based illusory correlations." Journal of Personality and Social Psychology, 60, 24-36.

Fiedler, K. (1996). Explaining and simulating judgment biases as an aggregation phenomenon in probabilistic, multiple-cue environments. Psychological Review, 103, 193-214.

Fiedler, K. (2000). Illusory correlations: A simple associative algorithm provides a convergent account of seemingly divergent paradigms. Review of General Psychology, 4, 25-58.

Fiedler, K. (2008). Pseudocontingencies: A key paradigm for understanding adaptive cognition. In N. Chater & M. Oaksford (Eds.), The probabilistic mind: Prospects for Bayesian cognitive science (pp. 165-186). London: Oxford University Press.

Fiedler, K., & Freytag, P. (2004). Pseudocontingencies. Journal of Personality and Social Psychology, 87, 453-467.

Fiedler, K., Freytag, P., & Meiser, T. (2009). Pseudocontingencies: An integrative account of an intriguing cognitive illusion. Psychological Review, 116, 187-206.

Fiedler, K., Freytag, P., & Unkelbach, C. (2007). Pseudocontingencies in a simulated classroom. Journal of Personality and Social Psychology, 92, 665-677. 

Fiedler, K., Freytag, P., & Unkelbach, C. (2011). Great oaks from giant acorns grow: How causal-impact judgments depend on the strength of a cause. European Journal of Social Psychology, 41, 162-172. doi: 10.1002/ejsp.750

Fiedler, K., & Graf, R. (1990). Grouping and categorization in judgment of contingency. In J. P. Caverni, J. M. Fabre, & M. Gonzalez (Eds.), Cognitive biases (pp. 47-57). Oxford, England: North-Holland.

Fiedler, K., Russer, S., & Gramm, K. (1993). Illusory correlations and memory performance. Journal of Experimental Social Psychology, 29, 111-136.

Fiedler, K., & Stroehm, W. (1986). The use of statistical, spatial-temporal, and intensional information in judgments of contingency. European Journal of Social Psychology, 16, 385-398.

Fiedler, K., Walther, E., Freytag, P., & Plessner, H. (2002). Judgment biases in a simulated classroom: A cognitive-environmental approach. Organizational Behavior and Human Decision Processes, 88, 527-561.

Fiedler, K., Walther, E., & Nickel, S. (1999). Covariation-based attribution: On the ability to assess multiple covariates of an effect. Personality and Social Psychology Bulletin, 25, 607-622.

Gordon, R. A. (1997). The moderation of distinctiveness-based illusory correlation: The impact of circadian variations and personal need for structure. Journal of Social Psychology, 137, 514-526.

Hamilton, D. L. (1981). Illusory correlations as a basis for stereotyping. In D. L. Hamilton (Ed.), Cognitive processes in stereotyping and intergroup behavior (pp. 115-144). Hillsdale, NJ: Erlbaum.

Hamilton, D. L., Dugan, P. M., & Trolier, T. K. (1985). The formation of stereotypic beliefs: Further evidence for distinctiveness-based illusory correlations. Journal of Personality and Social Psychology, 48, 5-17.

Hamilton, D. L., & Gifford, R. K. (1976). Illusory correlation in interpersonal perception: A cognitive basis of stereotypic judgments. Journal of Experimental Social Psychology, 12, 392-407.

Hamilton, D. L., & Sherman, S. J. (1989). Illusory correlations: Implications for stereotype theory and research. In D. Bar-Tal, C. F. Graumann, A. W. Kruglanski, & W. Stroebe (Eds.), Stereotyping and prejudice: Changing conceptions (pp. 59-82). New York: Springer-Verlag.

Hamilton, D. L., & Sherman, J. W. (1994). Stereotypes. In R. S. Wyer & T. K. Srull (Eds.), Handbook of social cognition (pp. 1-68). Hillsdale, NJ: Erlbaum.

Hamilton, D. L., Stroessner, S. J., & Mackie, D. M. (1993). The influence of affect on stereotyping: The case of illusory correlations. In D. M. Mackie & D. L. Hamilton (Eds.), Affect, cognition, and stereotyping: Interactive processes in group perception (pp. 39-61). New York: Academic Press.

Hammond, J. L. (1973). Two sources of error in ecological correlations. American Sociological Review, 38, 764-777.

Hermann, C., Ofer, J., & Flor, H. (2004). Covariation bias for ambiguous social stimuli in generalized social phobia. Journal of Abnormal Psychology, 113, 646-653.

Johnson, C., & Mullen, B. (1994). Evidence for the accessibility of paired distinctiveness in distinctiveness-based illusory correlation in stereotyping. Personality and Social Psychology Bulletin, 20, 65-70.

Johnson, C., Mullen, B., Carlson, D., & Southwick, S. (2001). The affective and memorial components of distinctiveness-based illusory correlations. British Journal of Social Psychology, 40, 337-358.

Kennedy, S. J., Rapee, R. M., & Mazurski, E. J. (1997). Covariation bias for phylogenetic versus ontogenetic fear-relevant stimuli. Behaviour Research & Therapy, 35, 415-422.

Leboeuf, R. A., & Norton, M. I. (2012). Consequence-cause matching: looking to the consequences of events to infer their causes. Journal of Consumer Research, 39, 128-141. doi: 10.1086/662372

McConnell, A. R., Sherman, S. J., & Hamilton, D. L. (1994). Illusory correlation in the perception of groups: An extension of the distinctiveness-based account. Journal of Personality and Social Psychology, 67, 414-429.

McGarty, C., Haslam, S. A., Turner, J. C., & Oakes, P. J. (1993). Illusory correlation as accentuation of actual intercategory difference: Evidence for the effect with minimal stimulus information. European Journal of Social Psychology, 23, 391-410.

Meiser, T. (2003). Effects of processing strategy on episodic memory and contingency learning in group stereotype formation. Social Cognition, 21, 121-156.

Meiser, T. (2006). Contingency learning and biased group impressions. In K. Fiedler & P. Juslin (Eds.), Information sampling and adaptive cognition (pp. 183-209). Cambridge, England: Cambridge University Press.

Meiser, T., & Broder, A. (2002). Memory for multidimensional source information. Journal of Experimental Psychology: Learning, Memory, and Cognition, 28, 116-137.

Meiser, T., & Hewstone, M. (2001). Crossed categorization effects on the formation of illusory correlations. European Journal of Social Psychology, 31, 443-466.

Meiser, T., & Hewstone, M. (2004). Cognitive processes in stereotype formation: The role of correct contingency learning for biased group judgments. Journal of Personality and Social Psychology, 87, 599-614.

Meiser, T., & Hewstone, M. (2006). Illusory and spurious correlations: Distinct phenomena or joint outcomes of exemplar-based category learning? European Journal of Social Psychology, 36, 315-336.

Mullen, B., & Johnson, C. (1990). Distinctiveness-based illusory correlations and stereotyping: A meta-analytic integration. British Journal of Social Psychology, 29, 11-28.

Pauli, P., Montoya, P., & Martz, G. E. (1996). Covariation bias in panic-prone individuals. Journal of Abnormal Psychology, 105, 658-662.

Pryor, J. B. (1986). The influence of different encoding sets upon the formation of illusory correlations and group impressions. Personality and Social Psychology Bulletin, 12, 216-226.

Risen, J. L., Gilovich, T., & Dunning, D. (2007). One-shot illusory correlations and stereotype formation. Personality and Social Psychology Bulletin, 33, 1492-1502.

Robinson, W. S. (1950). Ecological correlations and the behavior of individuals. American Sociological Review, 15, 351-357.

Sanbonmatsu, D. M, Shavitt, S., & Sherman, S. J. (1991). The role of personal relevance in the formation of distinctiveness-based illusory correlations. Personality and Social Psychology Bulletin, 17, 124-132.

Schaller, M., & Maass, A. (1989). Illusory correlation and social categorization: Toward an integration of motivational and cognitive factors in stereotype formation. Journal of Personality and Social Psychology, 56, 709-721.

Smith, E. R. (1991). Illusory correlation in a simulated exemplar-based memory. Journal of Experimental Social Psychology, 27, 107-123.

Spears, R., van der Plight, J., & Eiser, J. R. (1986). Generalizing the illusory correlation effect. Journal of Personality and Social Psychology, 51, 1127-1134.

Stroessner, S. J., Hamilton, D. L., & Mackie, D. M. (1992). Affect and stereotyping: The effect of induced mood on distinctiveness-based illusory correlations. Journal of Personality and Social Psychology, 62, 564-576.

Tomarken, A. J., Mineka, S., & Cook, M. (1989). Fear-relevant selective associations and covariation bias. Journal of Abnormal Psychology, 98, 381-394.

Tomarken, A. J., Sutton, S. K., & Mineka, S. (1995). Fear-relevant illusory correlations: What types of associations promote judgmental bias? Journal of Abnormal Psychology, 104, 312-326.

White, P. A. (1995). Use of prior beliefs in the assignment of causal roles: Causal powers versus regularity-based accounts. Memory & Cognition, 23, 243-254.

White, P. A. (2001). Causal judgment from contingency information: Relation between subjective reports and individual tendencies in judgment. Memory & Cognition, 28, 415-426.

Whitson, J. A., & Galinsky, A. D. (2008). Lacking control increases illusory pattern perception. Science, 322, 115-117.

Academic Scholar?
Join our team of writers.
Write a new opinion article,
a new Psyhclopedia article review
or update a current article.
Get recognition for it.

Last Update: 7/5/2016