Luckiness, competition, and performance on a psi task

This paper presents experimental data from an ongoing program of research into the psychology and parapsychology of luck and luckiness (see Smith, Wiseman, Harris, & Joiner, 1996). Part of this research explores the relationship between luck and psi by examining whether "lucky" people (i.e., individuals who report that seemingly chance events consistently tend to work out in their favor) outperform unlucky individuals on a variety of psi tasks.

In the past, researchers have studied this potential luck-psi relationship from many different perspectives. Greene (1960) investigated whether or not perceived luckiness correlated with psi performance. Participants were classified as lucky or unlucky by their responses on the "Greene Luck Questionnaire," which asked participants about whether or not they tended to win at games of chance, and were then asked to psychically influence the throwing of a ten-sided die. Greene found no relationship between luckiness and PK.

Ratte and Greene (1960) embedded a similar task in a game situation in which throws of the dice determined the outcome of different stages in an imaginary basketball game. Using a revised version of the Greene Luck Questionnaire, lucky participants tended to perform above chance while unlucky ones tended to perform below chance. Ratte (1960) went on to compare the performance of lucky and unlucky individuals in four conditions: game versus non-game and competitive versus non-competitive. Ratte reported that participants tended to perform better in the competitive condition than in the non-competitive condition, and the difference between the conditions tended to be due to the lucky rather than unlucky participants.(1)

More recently, Broughton (1979) carried out a pilot study in which he administered the Greene Luck Questionnaire to participants undertaking a computerized micro-PK task and found a significant positive correlation between luckiness and PK scores, although this correlation was nonsignificant in confirmatory data. Gissurarson and Morris (1991) found nonsignificant correlations between micro-PK performance and individuals' scores on either the Greene Luck Questionnaire or more general questions about everyday luck. Most recently, Rebman and Radin (1995) correlated computerized micro-PK performance with responses on a questionnaire concerning various aspects of luck. Nonsignificant negative correlations were found between psi scores and questions about how lucky participants considered themselves. However, significant negative correlations were reported between psi scores, participants' belief in luck, and whether or not they felt that they controlled their luck.

Other studies have examined perceived luckiness and ESP ability. For example, Rammohan and Krishna Rao (1987) administered multiple-choice questions to participants, the answers to which could only be known by ESP. They found no difference in ESP scores between participants who indicated that they tended to be lucky and those that tended to be unlucky in academic examinations. However, in a later study, Rammohan and Lakshmi (1993) found a significant positive correlation between psi scores and self-rated luckiness in exams on a nonintentional ESP test. Wiseman, Harris, and Middleton (1994) have also recently examined the possible relationship between self-reported luckiness and ESP performance. Perceived luckiness was measured by asking participants to rate how lucky they considered they had been in the past and expected to be in the future, in various aspects of their lives (e.g., health, relationships, games of chance, etc.). A significant positive correlation was found between perceived luckiness and ESP performance for participants who believed that the ESP task depended on non-chance factors. Perceived luckiness was not correlated with ESP performance for participants who believed that the outcome of the ESP task depended on chance.

The present study builds upon previous research in several ways. First, most of the above studies have employed the Greene Luck Questionnaire. However, this measures a rather limited notion of luck in that it is almost entirely restricted to success at gambling. The present study uses a new questionnaire which widens the definition of luck by asking people to classify themselves on the basis of seemingly chance events in everyday life (e.g., meeting the right person at the right time) as well as games of chance. On the basis of the cited past research, it was predicted that participants who described themselves in our questionnaire as lucky would perform better on the psi task than participants who described themselves as unlucky.

Second, this study examined the relationship between how well the subjects expected to do on the psi task and their perceived luckiness. It was expected that the self-described lucky participants would be more optimistic about their performance on the psi task than unlucky participants (Wiseman, et al, 1994).

Third, it was predicted that there would be a positive correlation between a subject's self-predicted psi performance and their actual psi performance.

Fourth, the study examined the possible role of competition in the potential luckiness-psi relationship. It was predicted that participants in the competitive condition would perform better on the psi task than participants in the non-competitive condition (Ratte, 1960).

METHOD

Design

This study used a mixed design. One between variable consisted of classifying participants' beliefs about their own luckiness into one of three categories: lucky, unlucky, or uncertain. The other between variable was competition, which had two levels, competitive or noncompetitive.

The two dependent variables consisted of participants' predictions concerning how well they would perform on the forthcoming PK task and their actual performance on the task.

Participants

Participants were 56 undergraduates at the University of Hertfordshire. Of these, 19 were male and 37 were female. Ages ranged from 18 to 50 years, with a mean age of 22.5 years. There were 30 participants in the competitive condition (12 males, 18 females; mean age 21.3), and 26 participants in the non-competitive condition (7 males, 19 females; mean age 23.7).

Apparatus/Materials

Luckiness Questionnaire (see Appendix). Participants were presented with two short descriptions: one of lucky people, and one of unlucky people. Lucky people were described as individuals for whom seemingly chance events tended to work out in their favor, while unlucky people were described as individuals for whom such events tend to work out against them. For each description, participants were asked to rate on a 7-point scale how well it described themselves, where 1 = doesn't describe me at all and 7 = describes me very well. Confidence ratings for each of these responses were obtained on another 7-point rating scale where 1 = not at all confident and 7 = extremely confident. Participants who gave the lucky description a rating of 4 or above, gave the unlucky description a lower rating, and had confidence ratings of at least 3 for both of these responses were categorized as lucky. Participants who gave the unlucky description a rating of 4 or above, gave the lucky description a lower rating, and had confidence ratings of at least 3 for both of these responses were categorized as unlucky. Participants who could not be classified as either lucky or unlucky due to, for example, identical ratings for the lucky and unlucky descriptions or low confidence ratings, were categorized as uncertain.

Predicted psi performance. Predicted psi performance was measured by asking participants to rate how well they thought they would do at correctly guessing the outcomes of coin-flips. Ratings were made on a 7-point rating scale where 1 = not at all well and 7 = extremely well.

Psi task. A measure of psi performance was made by asking participants to guess the outcomes of 50 computer-generated coin-flips. The task was programmed in HyperCard[TM] and presented on an Apple Macintosh computer (models varied between SE/30, LCIII, and Performa 475). For each coin-flip, participants are presented with a graphical representation of a boy who flips a coin and asks the participant to guess whether or not the coin has landed heads or tails (participants cannot see whether the coin has landed heads or tails). Participants make a guess by clicking the mouse on a heads box or a tails box. After making a guess, participants are immediately told the outcome of the coin-flip and whether or not he/she was correct. This is repeated for 50 coin-flips. The outcome of each coin-flip is based upon three samplings from the computer's own pseudo-RNG taken at the start of each trial (however, the RNG was seeded only once when the computer was turned on); each sampling produces a number between I and 100. The third sampling of the RNG decides which of the first two samplings is actually used to determine the outcome of the coin-flip (a number from 1-50 means the first sampling is used, a number from 51-100 meant the second sampling would be used). A resulting number from 1-50 translated into heads, a number from 51-100 translated into tails. Thus if, for example, the three samplings were 23, 81, and 56, the outcome of the coin-flip would be tails (56 would designate 81 as the number to determine the coin-flip, which would translate into tails). The pseudo-RNG output produced an unbiased distribution of heads and tails, and no obvious sequential dependencies (i.e., lag one) were found.

In the competitive condition, a competition score chart was used to record who was winning at each of five stages in the "game," and sweets (candy) were used as prizes for the winner of each stage.

Procedure

Participants in the non-competitive condition were tested in two groups (11 participants in one, 15 in the other) on two consecutive days as part of a research methodology class. In each group participants were tested simultaneously. Each participant sat in front of an Apple Macintosh computer. They were told that they were about to take part in an experiment on the psychology and parapsychology of luck, and were asked to complete our Luckiness Questionnaire. They were then told that they were about be asked to guess the outcomes of a series of computer-generated coin-flips (they were not told how many coin-flips there would be) which would be randomly determined by the computer and presented on the screen. Their aim was to try to correctly guess as many coin-flips as they could. For each coin-flip, they were to make their guess by clicking the mouse on the appropriate box. Before beginning this task, they were each asked to rate how well they thought they would do at correctly guessing the outcome of the coin-flips on a 7-point rating scale. They then started the program, and when 50 coin-flips had been completed, the program asked participants to record their age and gender.

The procedure for the competitive condition was the same as above, apart from the fact that participants were tested in pairs (each pair consisting of two friends), and were instructed to try to beat their friend at correctly guessing the outcomes of the coin-flips. The element of competition was enhanced by recording who was winning at each of five stages; each stage consisted of ten coin-flips, although participants were not told this nor were they told how many stages there would be. For each stage, one participant within each pairing completed ten coin-flips, then the other participant completed ten coin-flips. After each stage, the person who had correctly guessed the most coin-flips in that stage was awarded a small sweet (a piece of candy). The order of play alternated between stages, and as in the non-competitive condition, participants were not told how many coin-flips they would be asked to complete in total.

All participants were thanked for their participation and fully debriefed.

RESULTS

Luckiness, competition and psi performance

The percentages of coin-flips guessed correctly by lucky, unlucky, and uncertain participants in the competitive and non-competitive conditions are given in Table 1.

TABLE 1

PERCENTAGES (MCE = 50%) OF COIN-FLIPS GUESSED CORRECTLY BY LUCKY,
UNLUCKY, AND UNCERTAIN PARTICIPANTS IN COMPETITIVE AND
NON-COMPETITIVE CONDITIONS

                                   Luckiness
                   unlucky   uncertain   lucky    Totals

competitive         50.40     50.33      53.00     51.07
                   (n = 10)   (n = 12)   (n = 8)   (n = 30)

non-competitive     48.67     44.80      50.44     48.54
                    (n = 12)  (n = 5)   (n = 9)    (n = 26)

Totals              49.46     48.71      51.65
                   (n = 22)   (n = 17)  (n = 17)

It would appear that analyzing the various aspects of luck and luckiness may lead to a fuller understanding of psi. The present study did not find evidence to support the hypothesis that lucky people perform better than unlucky people on tests of psi ability. However, continued research in this area could potentially cast light on how psi is conceptualized in people's minds, and may even make a significant contribution to the search for a reliable indicator of successful psi performance.

The authors would like to acknowledge the financial support of the Leverhulme Trust, London, for funding the research reported in this paper.

1 It should be noted that Stanford (1977) has argued that the statistical analyses used by Ratte (1960) and Ratte and Greene (1960) were inappropriate.

REFERENCES

BROUGHTON, R. S. (1979). An experiment with the Head of Jut. European Journal of Parapsychology, 2, 337-357.

GISSURARSON, L. R., & MORRIS, R. L. (1991). Examination of six questionnaires as predictors of psychokinesis performance. Journal of Parapsychology, 55, 119-145.

GREENE, F. M. (1960). The feeling of luck and its effect on PK. Journal of Parapsychology, 24, 129-141.

PALMER, J. (1972). Scoring in ESP tests as a function of belief in ESE Part II: Beyond the sheep-goat effect. Journal of the American Society for Psychical search, 66, 1-26.

RAMMOHAN, V. G., & KRISHNA RAO, E V. (1987). ESP performance in classroom and examination settings. Journal of Indian Psychology, 6, 24-37.

RAMMOHAN, V. G, & LAKSHMI, V. V. (1993). A comparative study of ESP under intentional vs. nonintentional conditions using the Rao-OBrien paradigm. Journal of Indian Psychology, 11, 24-31.

RATTE, R.J. (1960). Comparison of game and standard PK testing techniques under competitive and noncompetitive conditions. Journal of Parapsychology, 24, 235-244.

RATTE, R. J., & GREENE, E M. (1960). An exploratory investigation of PK in a game situation. Journal of Parapsychology, 24, 159-170.

REBMAN, J. M., & RADIN, D. I. (1995). A preliminary analysis of luck and psi. Proceedings of Presented Papers: The Parapsychological Association 38th Annual Convention, 409-417.

SMITH, M.D., WISEMAN, R., HARRIS, P., & JOINER, R. (1996). On being lucky: The psychology and parapsychology of luck. European Journal of Parapsychology, 19, 35-43.

STANFORD, R. G. (1977). Experimental psychokinesis: A review from diverse perspectives. In B. B. Wolman (ed.), Handbook of Parapsychology, (pp. 324381). New York: Van Nostrand Reinhold Company.

STANFORD, R. G. (1990). An experimentally testable model for spontaneous psi events: A review of related evidence and concepts from parapsychology and other sciences. In S. Krippner (Ed.), Advances in Parapsychological Research (pp. 54-167). NC: McFarland.

WAGENAAR, W. A., & KEREN, G. B. (1988). Chance and luck are not the same. Journal of Behavioural Decision Making, 1, 65-75.

WISEMAN, R., HARRIS, P., & MIDDLETON, W. (1994). Luckiness and psi: An initial study. Journal of the Society for Psychical Research, 60, 1-15.

Perrott-Warrick Research Unit Department of Psychology University of Hertfordshire College Lane Hatfield AL10 9AB UK

University of Sussex Open University

APPENDIX

LUCKINESS QUESTIONNAIRE

People consider themselves lucky or unlucky for a wide variety of reasons. At this stage in our research, we are interested in whether people consider themselves lucky or unlucky when it comes to seemingly chance events.

Please read each of the descriptions that follow, and, for each one, rate how well it describes you using the scale below, by writing the number that best represents your response in the space provided:

Doesn't describe                                 Describes me
me at all                                        very well

1                  2     3     4     5     6         7

COPYRIGHT 1997 Parapsychology Press
COPYRIGHT 2004 Gale Group