The effect of male confederate presence, betting, and accuracy of play on males' gambling on blackjack.
Abstract: Previous research suggests that the actions of a confederate can alter participants' gambling behavior. In the present experiment, male participants played Blackjack either alone or in the presence of a confederate. The confederate either quit early in the session or played for the entire session. Across sessions in which the confederate played for the entire session, how much the confederate bet per hand and how accurately he played were manipulated. Participants gambled significantly more money across the session when a confederate played the entire session than when a confederate left early in the session. During sessions in which the confederate played for the entire session, the number of hands participants played and their total amount bet varied directly with the confederate's bet size. Overall, the results indicate that gambling behavior can be indirectly influenced by another's gambling behavior, which may provide useful data for researchers and therapists seeking to predict and control gambling behavior.

Key words: Gambling, blackjack, bet size, confederates, males
Article Type: Report
Subject: Gambling (Research)
Gambling (Psychological aspects)
Blackjack (Game)
Authors: McDougall, Casey L.
Terrance, Cheryl
Weatherly, Jeffrey N.
Pub Date: 06/22/2011
Publication: Name: The Psychological Record Publisher: The Psychological Record Audience: Academic Format: Magazine/Journal Subject: Psychology and mental health Copyright: COPYRIGHT 2011 The Psychological Record ISSN: 0033-2933
Issue: Date: Summer, 2011 Source Volume: 61 Source Issue: 3
Topic: Event Code: 310 Science & research
Product: Product Code: 7760000 Legal Gambling NAICS Code: 7132 Gambling Industries
Accession Number: 265302080
Full Text: Gambling is rarely studied experimentally. An all-fields search for "gambling" using PsycINFO (conducted on January 27, 2010) yielded a total of 4,506 articles. However, when "gambling" was paired with the term "experiment," only 214 articles were identified, suggesting that as little as 5% of the gambling literature involves experimentation. This low percentage is discouraging because the experiment represents the most straightforward and potent way to determine cause-and-effect relationships. On the other hand, at least some experimental work is being done in the area (e.g., Brandt & Pietras, 2008; Dixon & Schreiber, 2004).

It is interesting to note that a significant number of experimental studies on gambling were conducted several decades ago (e.g., Blascovich & Ginsburg, 1974; Blascovich, Ginsberg, & Howe, 1975; Blascovich, Ginsburg, & Veach, 1975; Blascovich, Veach, & Ginsburg, 1973), with many of these investigations designed to study the phenomenon of the "risky shift." Overall, the results of these studies indicate that social variables (e.g., the presence and behavior of other gamblers) can influence gambling behavior. For instance, Blascovich, Ginsberg, and Howe (1975) found that participants playing the card game Blackjack would bet more credits when in groups of three than when playing alone, a result that replicated that of other studies that had employed slightly different procedures (Blascovich et al., 1973; Blascovich, Ginsburg, & Veach, 1975). Blascovich and Ginsburg (1974) employed confederates to play Blackjack along with the participants, with the confederates consistently betting a low ($.10), moderate ($30), or high ($.50) amount. Their results showed that participants' betting was influenced by the wagering of the confederates but that wagering tended to increase across time for all participants, even those playing alone.

Our laboratory has also experimentally investigated how gambling can be influenced by others. For instance, McDougall, McDonald, and Weatherly (2008) investigated whether the gambling of American Indian and non-Indian participants would be sensitive to the actions and/or ethnicity of another gambler (i.e., a confederate) when playing a slot-machine simulation. Eight male American Indians and eight male non-Indians participated in five gambling sessions. In one session, the participant gambled alone. In the other four, the participant played in the presence of one confederate who was of the same or different ethnicity and who gambled for the entire session or quit the session after playing the slot-machine simulation five times. Gambling of the participants did not vary by ethnicity or by the ethnicity of the confederate. However, gambling behavior was altered by the confederate's actions, with participants playing the slot machine fewer times and betting fewer credits across the session when the confederate left the session than when alone or when the confederate stayed and gambled. Results did not, however, show that gambling behavior increased when the confederate stayed and gambled relative to when the participant gambled alone.

Thus, Blascovich and colleagues reported finding that the presence of other gamblers facilitated gambling, a result that was not replicated by McDougall et al. (2008). However, the McDougall et al. study employed a slot-machine simulation, whereas the previous research had participants betting on Blackjack. But there were other procedural differences that may have produced these some-what discrepant results. Briefly stated, across the studies by Blascovich and colleagues, one could argue that the findings were influenced by one or more of the following: order effects, the participants' not wagering for real money (i.e., getting to keep their winnings), the participants' inability to voluntarily cease gambling, or controlling for the behavior of the "other" gamblers. McDougall et al.'s procedure (2008) was designed to control for these issues. Thus, the fact that the results of McDougall et al. (2008) differed from previous studies may have been the result of one or more of these differences.

The present study was designed with several goals in mind. The first was to extend the findings of McDougall et al. (2008) to a different gambling game. The second was to reconcile the results of McDougall et al. with those of previous studies (e.g., Blascovich & Ginsburg, 1974) that have found that the behavior of others can increase gambling behavior. The third was to make a novel contribution to the gambling literature. That is, an attempt was made to determine whether how accurately other players played a game would influence the gambling behavior of the participant.

To accomplish these goals, male university students were recruited to play Blackjack in a controlled casino-laboratory environment. Participants played in six distinct gambling sessions, and several variables were manipulated across those sessions. The first was whether a second gambler (a confederate) was present during the session. The second was whether the confederate, when present, quit playing early in the session or continued to play for the entire session. The third was how the confederate bet during the sessions in which he stayed for the entire session. The fourth was how the confederate played his cards during the sessions in which he stayed for the entire session.

On the basis of previous research and the results of McDougall et al. (2008), we predicted that participants' gambling behavior would be influenced by the presence of another gambler. Specifically, we predicted that participants would play fewer hands and bet less money when the confederate left the session than when the confederate stayed and played for the entire session or was absent altogether. Consistent with prior research (e.g., Blascovich & Ginsburg, 1974), however, we also predicted that participants would play more hands and bet more money when the confederate played the entire session than they would when the confederate was absent.

In terms of the behavior of the confederate, we predicted that the manipulation of the confederate's bet size would influence the participants' gambling behavior based on the results of Blascovich and Ginsburg (1974). However, the literature is silent on how the confederate's accuracy of play would influence gambling behavior. This manipulation was also less overt than that of bet size (e.g., the participant might or might not recognize an inaccurate play when it is made). Thus, although we predicted that the participants' gambling would be influenced by the confederate's accuracy of play, we were less confident about this prediction than about the others.

Method

Participants

The present study used a total of 16 male participants. The participants were recruited through the psychology department's participant pool at the University of North Dakota, To participate, each individual was required to be at least 21 years of age and score less than 5 on the South Oaks Gambling Screen (SOGS; Lesieur & Blume, 1987). The mean age of the participants was 23.20 years (SD = 2.81 years). The mean SOGS score was 1.25 (SD = 1.39).

Participation was limited to males for several reasons. First, the literature indicates that males display higher instances of problematic gambling behavior than females (see Petry, 2005). Second, research historically suggests that males tend to prefer table games when gambling, whereas females tend to prefer slot machines or bingo (see LaPlante, Nelson, LaBrie, & Shaffer, 2006, for a discussion). Third, adding gender as a pseudo-independent variable would have greatly increased the number of necessary sessions (because gender of the confederate would also need to be taken into account), which would have made completion of the present procedure infeasible. Participants were compensated with extra course credit for their participation, as well as whatever amount of money they had won or had remaining across the six Blackjack sessions.

Apparatus

The laboratory casino was furnished as realistically as possible. It included an authentic Blackjack table and stools typically found in an American casino. A "No tipping" sign was located on the Blackjack table. For the Blackjack game proper, there were regulation gaming chips, four decks of playing cards, a dealer's "shoe" to hold the cards, and a professional Blackjack dealer. The casino was located in a room that measured approximately 4 m by 5 m.

Materials

The materials included the SOGS, which consists of 20 questions designed to assess the level of a person's gambling experience. Potential problematic gambling behavior is noted by scores of 3 or 4, while pathological gambling may be present if the person scores 5 or more. Those participants scoring 5 or more were immediately given their extra course credit and dismissed. Two potential participants scored 5 or more on the SOGS; they were replaced.

A second questionnaire was a brief quiz used to assess the participant's knowledge of Blackjack. The quiz consisted of 15 items addressing Blackjack play and certain gambling fallacies. This quiz was administered twice and served to help us monitor changes in participants' knowledge of the game due to participating in the six sessions. The list of the questions in this quiz can be found in the Appendix.

Procedure

Participants were recruited via a sign-up sheet in the psychology department. At the beginning of the first session, the researcher checked the participant's identification to ensure that he was at least 21 years of age. Next, the researcher went through the process of obtaining informed consent as approved by the Institutional Review Board at the University of North Dakota. Subsequently, the participant completed the SOGS. The participant then completed the Blackjack quiz, during which time the researcher scored the SOGS to determine whether he could continue participation. If the participant's first session included a confederate, the researcher informed the participant that the confederate had completed these measures in a previous session.

The participant was then seated at the Blackjack table. If a confederate was present, the players were assigned Seat 1 or 2 (which determined who received the first card during game play). This order was counterbalanced across sessions and participants for sessions in which a confederate was present. The researcher then read the following instructions:

Participants' questions were answered by repeating the instructions. After the researcher had read the instructions and answered any questions, the participant then played the game until one of the three criteria for ending the session was met. At the end of the sixth session, the participant completed the Blackjack quiz for the second time. The researcher then read a debriefing statement to the participant, paid the participant for the total value of gaming chips he had accrued across the six sessions, and dismissed the participant from the study.

The conditions in the six sessions are depicted in Table 1. It shows whether a confederate was present (5 sessions) or absent (1 session). In all sessions in which a confederate was present, he was treated identically to the participant (with the exception of the first session's questionnaires). In two of the four sessions in which the confederate was present and played the entire session, the confederate was instructed to play "accurately" as determined by the basic strategy for Blackjack proposed by Braun (1980). In the other two sessions, the confederate was instructed to deviate from this strategy, on average, every other hand. This "on average" procedure was instituted to alleviate the problem of the confederate overtly advertising the independent variable by taking a hit card on a hand in which doing so was blatantly unreasonable (e.g., taking a hit when holding a hand with the value of 20 when then goal is to get as close to the value of 21 as possible without going over 21). Furthermore, in two of the sessions in which the confederate was present and played the entire session, the confederate bet the minimum allowable bet each hand (i.e., $.10). In the other two sessions, he bet the maximum allowable bet each hand (i.e., $.50). These bet sizes were chosen because they matched those of Blascovich and Ginsburg (1974).

In the one session in which the confederate was present but played only briefly, the confederate was instructed to leave after he won for the first time. In the meantime, he was instructed to play every three of four hands accurately (i.e., 75% accuracy). This accuracy rate was chosen because it equaled the rate displayed across the four confederate "stay" sessions. Furthermore, the confederate bet $.30 per hand, which was the average bet size across the four confederate "stay" sessions.

Participants completed a maximum of one session per calendar day. The order of the six different sessions was counterbalanced across participants. A total of two confederates were employed, so participants did not necessarily play every session with the same confederate. The researcher was present during every session to record the number of hands played, bet per hand, outcome of each hand, and accuracy of play of the confederate. In terms of interactions between the participant and the confederate/dealer, both the confederates and the dealer were instructed to respond "normally" to interactions initiated by the participant, but not to overtly initiate verbal interactions with the participant. If the participant asked the dealer's advice about a certain hand, the dealer was instructed to respond accurately; however, he was instructed not to offer advice unless prompted. Although specific data on when participants asked advice were not recorded, such instances were rare. Phrased differently, participants received no formal guidance within the procedure on how to play Blackjack.

Dependent Measures

Several dependent measures were of interest in the present study. One was the total number of hands played per session, which can be considered a measure of persistence. Another was the total amount of money bet per session, which can be considered a measure of risk. These two variables allowed for the calculation of a third variable, average bet size per hand. This measure was potentially useful because it was legitimately possible that participants could play more hands in the alone session than when the confederate was present simply because there was only one player for the dealer to deal to.

Another dependent measure was the amount of money the participants had remaining at the end of the session. This measure was analyzed to determine if the participants' experienced outcomes, rather than the independent variables, might have influenced the results. Next, the scores on the two Blackjack quizzes were analyzed to determine if participants' knowledge of the game changed over the course of the experiment. Next, the accuracy of play of the confederates was measured to verify that the independent variable was manipulated as described. Lastly, the accuracy of the researcher's observations was verified through interobserver agreement. Multiple researchers observed play in approximately 5% of the sessions and independently recorded the cards dealt, bet sizes, outcomes, and confederate accuracy. Across all measures in these sessions, interobserver agreement was high ([lambda] =.88).

Results

Presence and Actions of the Confederate

To determine the effect of the presence and actions (i.e., leaving or staying) of the confederate, the mean number of hands played, total amount of money bet, and average bet per hand were calculated across the four sessions in which the confederate stayed and played the entire session. These means were then compared to the same variables for the individual sessions in which the participant played alone and when the confederate was present but left the session early. Figure 1 presents the mean number of hands played, total amount of money bet, and bet size per hand for the alone session, the confederate "leaves" sessions, and the average across the four confederate "stays" sessions. The number of hands participants played per condition was analyzed by conducting a one-way repeated measures ANOVA. Results indicated a significant difference in the number of hands played per session, F (2, 30) = 20.87, p < .001, [[eta].sup.2] =.582. A Tukey HSD post hoc test showed that participants played fewer hands in the leave session than in either the alone session or the stay sessions. The number of hands in the alone session and the stay sessions did not differ significantly. Results from these analysis, and all that follow, were considered significant at p < .05.

[FIGURE 1 OMITTED]

A one-way repeated measures ANOVA conducted on the total amount of money bet across these three conditions also produced a significant effect, F (2, 30) = 12.44, p < .001, [[eta].sup.2] =.453. A Tukey HSD post hoc test showed that participants bet significantly less money in the alone session than in the stay sessions. They also bet significantly less money in the leave session than in the stay sessions. However, the total amount bet in the alone and leave sessions did not differ significantly. A one-way repeated measures ANOVA was conducted on the average bet per hand across these three conditions. This analysis did not produce a significant effect, F(2, 30) = 1.48, p =.245, [eta] =.089. Thus, the effect of the presence and actions (i.e., leaving or staying) of the confederate did not influence the size of the participants' bets on individual hands.

The final analysis was a one-way repeated measures ANOVA conducted on the amount of money participants had remaining at the end of the alone session, confederate leaves sessions, and average of the confederate stays sessions. That analyses yielded a significant effect, F(2, 30) = 6.48, p =.005, [eta] =.302. Post hoc tests indicated that participants ended the session with significantly more money in the sessions in which the confederate stayed (M = $6.31, SD = $1.77) than they did either in the sessions when they played alone (M = $4.97, SD = $1.78) or when the confederate left (M = $4.51, SD = $1.99). There was no significant difference in money remaining between the alone and confederate leaves sessions.

Accuracy of Play and Bet Size

Confederates played accurately on 74% of hands in the "leave" sessions, on 99% of the hands in the "stay" sessions in which they were instructed to play accurately, and on 53.5% of the hands in the "stay" sessions in which they were instructed to play inaccurately. Additionally, the confederate played the entire session in the "stay" sessions (i.e., did not lose all of his chips) in 93.75% of the sessions.

Figure 2 presents the mean number of hands played, total amount of money bet, and bet size per hand when the confederate played accurately and inaccurately as a function of the confederate's bet size. All data in Figure 2 come from the sessions in which the confederate stayed and played. The effect of confederates' accuracy of play and bet size was analyzed by conducting separate two-way (Accuracy of play x Bet size) repeated measures ANOVAs on the number of hands played per session, total amount of money bet, and bet size per hand measures. Analysis of the number of hands played failed to yield a significant main effect of accuracy of play, F < 1, [[eta].sup.2] =.003, or main effect of bet size, F < 1, [[eta].sup.2] = -040. The interaction between accuracy of play and bet size was also not significant, F < 1, [[eta].sup.2] =.013.

[FIGURE 2 OMITTED]

Analysis of the total amount bet per session yielded a nonsignificant main effect of accuracy of play, F < 1, [[eta].sup.2] =.046. However, the main effect of bet size was significant, F(l, 15) = 9.76, p =.007, [[eta].sup.2] =.395, indicating that participants bet more money when the confederate placed the maximum bet than when the confederate placed the minimum bet. The interaction between accuracy of play and bet size was not significant, F < 1, if =.056-

Analysis of the bet-per-hand data again yielded a nonsignificant main effect of accuracy of play, F < 1, [[eta].sup.2] =.007. The main effect of bet size was significant, F(l, 15) = 20.36, p < .001, [[eta].sup.2] =.576, with participants betting significantly more per hand when the confederate always placed the maximum bet than when the confederate always placed the minimum bet. Lastly, the interaction term was not significant, F < 1, [[eta].sup.2] =.026.

A one-way repeated measures ANOVA conducted on the amount of money participants had remaining at the end of each of the four stay conditions did not reveal a significant effect, F(3, 45) = 1.84, p =.154, [[eta].sup.2] =.109, indicating that participants ended each type of session with a similar amount of money.

On average, the participants scored 3 points higher on the postexperiment Blackjack quiz than they did on the pre-experiment quiz, A one-way repeated measures ANOVA was conducted on participants' pre-and postquiz scores. The effect was significant, F (l, 15) = 15.99, p <.001, [[eta].sup.2] =.516. This increase in knowledge did not influence the above results, however. When the above measures were analyzed as a function of session order (i.e., disregarding condition), significant results were not observed. Thus, participation increased participants' knowledge of the game of Blackjack, but this change in knowledge appeared to be independent of the effect of the confederate's presence, actions, and bet size.

Discussion

The present study was designed to test whether participants' gambling would be influenced by the presence of another gambler. Specifically, would participants' gambling behavior be decreased by another gambler being present but leaving the session early, as in McDougall et al. (2008), and/or be increased by another gambler being present and gambling for the entire session, as in several studies by Blascovich and colleagues (e.g., Blascovich, Ginsburg, & Howe, 1975)? Results showed a significant effect of the confederate being present during the session and leaving or staying. Participants played significantly fewer hands when the confederate was present but left the session early than they did when alone or when the confederate was present and played for the entire session. Participants bet significantly more money across the session when the confederate was present and played for the entire session than when alone or when the confederate was present but left the session early.

The present study was also designed to determine whether how the confederate played during the session would influence the participants' gambling behavior. The factors manipulated when the confederate was present and played for the entire session were how much the confederate bet each hand and whether the confederate played the hands accurately or inaccurately. Bet size influenced participants' behavior, with participants risking more money when the confederates placed the maximum bet than when the confederate placed the minimum bet. Participants' gambling was not influenced by the accuracy of the confederate's play.

Finding that participants decreased the number of hands they played when the confederate was present but left the session early replicates the finding of McDougall et al. (2008) and extends it to Blackjack, rather than slot-machine, play. Finding that participants increased the amount of money they bet when the confederate was present and played for the entire session replicates the findings from a series of studies in the 1970s by Blascovich and colleagues (e.g., Blascovich, Ginsburg, & Howe, 1975). However, it is important to note that, on average, participants ended the "stay" sessions with more money than they did in the other sessions (and more than they were staked; i.e., they won). Thus, this monetary difference could potentially explain the effect of the confederate staying. It is also worthy of mention that the measure of bet per hand did not differ across these conditions. Thus, in the present study, the effect of the confederate being present and leaving/staying was one of how many hands were played. This outcome was not linked to the possible speed of play. It was possible for more hands to be played during a session when the confederate was absent or left than when he was present for the entire session. With that said, the exact timing of hands was not recorded in the present study; thus, it is not possible to determine whether the decrease in the number of hands played was a function of quitting earlier in the session or increases in the time participants took to play each hand. At a theoretical level, one could potentially interpret the present finding as an increase in the participants' propensity to gamble. However, it is also possible that it represents a social-facilitation effect in which behavior was influenced by the presence of another person (Triplett, 1898; Zajonc, 1965; see Guerin, 1993, for a review). Subsequent research is needed to determine which, if either, of these explanations is correct.

Whether the confederate played accurately or inaccurately did not influence the participants' gambling behavior. It is possible that the failure to observe a significant effect was one of statistical power. However, even the largest effect size observed for this particular variable was small ([[eta].sup.2] =.046; Cohen, 1988), meaning that a substantial increase in the number of participants would have been required for this manipulation to approach statistical significance. Several potential reasons exist for why no effect was observed. One is the range of manipulation (e.g., confederates could have played perfectly accurately in some conditions and completely inaccurately in others). Another was the possibility that the manipulation was too subtle, which could have been avoided by having the dealer overtly point out to the confederate (and thus the participant) instances in which the confederate played inaccurately. It is also the case that participants' knowledge of Blackjack increased across the experiment, suggesting that they were relatively novice players at the outset and thus potentially unlikely to know whether or not the confederate was playing accurately.

Consistent with the results of Blascovich and Ginsburg (1974), the present results showed that the size of the bet placed by the confederate did influence the participants' gambling. This effect, unlike the effect of the confederate's presence, was one of risk. That is, participants played a similar number of hands regardless of the confederate's bet size. However, the total amount of money participants risked and their average bet size were increased when the confederate always bet the maximum amount.

The question is, why was this outcome observed? It is possible that the participants were displaying a conformity effect (e.g., Asch, 1955, 1956; Sherif, 1936). If the outcome is linked to conformity, then it is worthy of note that such an effect was observed with only one confederate. Studies on conformity typically find that a group of confederates is required to show an effect. Another possibility is that it was the outcome of competition (e.g., Wilson & Daly, 1985), although we did not attempt to create a competitive environment. Still another, perhaps related, possibility is that participants were trying to avoid a negative social stigma. Whatever the reason, it is clear that there is a social contingency present, even in a laboratory situation, that can facilitate gambling behavior. Thus, given the prevalence of gambling problems and the fact that many gambling situations take place in a social context, further examination of this particular influence is particularly warranted.

One positive implication that can be taken from the present results is that not only can the presence and actions of others facilitate gambling, but they can also inhibit it. Even more encouraging is the finding that this effect is observed with only one other gambler. This result might help explain why people suffering from behavioral disorders, such as pathological gambling, experience better treatment outcomes when they have an established social support system than when they do not (e.g., Korn & Shaffer, 1999). The present results suggest that moderate gambling might be promoted by fellow gamblers who gamble for only a short period of time and refrain from placing the maximum bets. Likewise, therapists would be well served to advise their clients against surrounding themselves with gamblers who gamble for lengthy periods at a time and who play for high stakes. The large effect sizes observed for these factors in the present study suggest that they are certainly worthy of consideration in treatment settings.

The present study does suffer from a number of potential limitations. One is that we had a relatively small, homogenous sample of nonpatho-logical male gamblers. Had we employed a nonuniversity sample of varying ages, backgrounds, and socioeconomic levels, it is possible that different results would have been observed. With that said, it is also possible that even larger effects would have been observed than was the case with the present sample. We avoided using pathological gamblers to ensure that such individuals were not allowed to engage in their pathology. However, it is possible that as pathological gamblers, these individuals would be less prone than nonpathological gamblers to be influenced by other gamblers.

One other potential limitation is the fact that participants in the present study were gambling with staked money. Although this limitation cannot be completely countered, previous studies from our laboratory have demonstrated that participants in gambling experiments play more conservatively when they are playing with staked money than they do when they are playing for hypothetical winnings (Weatherly & Brandt, 2004; Weatherly & Meier, 2007), indicating that staked money does hold value for the participants. Likewise, research on the "endowment effect" (Kahneman, Knetsch, & Thaler, 1990) indicates that when individuals are gifted with something, they psychologically "own" it and attempt to avoid its loss. Other research, however, indicates that college students place less value on "won" money than they do on their own money (Weatherly, Derenne, & Terrell, 2010).

Finally, one could also argue that the stakes for which participants were playing were not high and do not represent the levels of risks taken by casino-going gamblers. This concern would seem legitimate, and the present data cannot address it. Future research might examine how the present effects vary as a function of the stakes involved.

Appendix

Blackjack Knowledge Survey

1. What should you do when you have ace-five and the dealer has a two?

a. hit b. stand c. double-down d. surrender

2. What should you do when you have three-seven and the dealer has a jack?

a. hit b. stand c. double-down d. surrender

3 What should you do when you have twelve and the dealer has a two?

a. hit b. stand c. double-down d. surrender

4 What should you do when you have king-queen and the dealer has a six?

a. hit b. stand c double-down d. surrender

5 When should you split fives?

a. Dealer has five or six

b. Dealer has four, five, or six

c. Dealer, has two, three, four, five, or six

d. Never

6. When should you split eights?

a. Dealer has two, three, four, five, or six

b. Dealer has two, three, four, five, six, seven, or eight

c. Always

d. Never

7. What should you do with a pair of threes if the dealer has a six?

a. hit b. stand c. double-down d. split pair

8. When should you take insurance?

a. When the dealer offers you insurance

b. When the dealer offers you even money

c. When you have a good hand (twenty or Blackjack)

d Never

9. Which of the following hands should you NOT surrender?

a. Nine-seven against dealer's ace

b. Jack-five against dealer's queen

c. Eight-six against dealer's ace

d. Six-ten against dealer's nine

10. What should you do with soft 18 when the dealer has a queen?

a. hit b. stand c. double-down d. surrender

11 What should you do with a pair of nines when the dealer has a seven?

a. hit b. stand c. split pair d. surrender

12 What should you do with a pair of fours when the dealer has a four?

a. hit b. stand c. double-down d. split pair

13 Which of the following rule variations favors the house?

a. Dealer must hit soft 17

b. Double-down after splitting

c. Re-splitting aces permitted (up to 4 hands only)

d. Early surrender

14. Which rule most favors the player?

a. Doubling down b. Splitting pairs c. Insurance pays 2:1

d. Blackjack pays 3:2

15. Can the dealer or other players influence the outcome of each hand of cards?

a. Yes b. No

References

ASCH, S. (1955). Opinions and social pressure. Scientific American, 193, 31-35.

ASCH, S. E. (1956). Studies of independence and conformity: I. A minority of one against a unanimous majority. Psychological Monographs, 70, 70.

BLASCOVICH, J., & GINSBURG, G. P. (1974). Emergent norms and choice shifts involving risk. Sociometry, 37, 205-218.

BLASCOVICH, J., GINSBURG, G. P., & HOWE, R. C. (1975). Blackjack and the risky shift: II. Monetary stakes. Journal of Experimental Social Psychology, 11, 224-232.

BLASCOVICH, J., GINSBURG, G. P., & VEACH, T. L. (1975). A pluralistic explanation of choice shifts on the risk dimension. Journal of Personality and Social Psychology, 3, 422-429. BLASCOVICH, J., VEACH, T. L., & GINSBURG, G. P. (1973). Blackjack and the risky shift. Sociometry, 36, 42-55.

BRANDT, A. E., & PIETRAS, C. J. (2008). Gambling on a simulated slot machine under conditions of repeated play. The Psychological Record, 58, 405-426.

BRAUN, J. H. (1980). How to play winning blackjack. Chicago: Data House.

COHEN, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum Associates.

DIXON, M. R., & SCHREIBER, J. E. (2004). Near-rniss effects on response latencies and win estimations of slot machine players. The Psychological Record, 54, 335-348.

GUERIN, B. (1993). Social facilitation. Cambridge, UK: Cambridge University Press.

KAHNEMAN, D., KNETSCH, J. L., & THALER, R. H. (1990). Experimental tests of the endowment effect and the Coase theorem. Journal of Political Economy, 98, 1325-1348.

KORN, D. A., & SHAFFER, H. J. (1999). Gambling and the health of the public: Adopting a public health perspective. Journal of Gambling Studies, 15, 289-365.

LAPLANTE, D. A., NELSON, S. E., LABRIE, R. A., & SHAFFER, H. J. (2006). Men and women playing games: Gender and the gambling preferences of Iowa gambling treatment program participants. Journal of Gambling Studies, 22, 65-80.

LESIEUR, H. R., & BLUME, S. B. (1987). The South Oakes Gambling Screen (SOGS): A new instrument for the identification of pathological gamblers. American Journal of Psychiatry, 144, 1184-1188.

MCDOUGALL, C. L., MCDONALD, J., & WEATHERLY, J. N. (2008). The gambling behavior of American Indian and non-Indian participants: Effects of the actions and ethnicity of a confederate. American Indian & Alaskan Native Mental Health Research, 14(3), 59-74.

PETRY, N. M. (2005) Pathological gambling: Etiology, comorbidity, and treatment. Washington, DC: American Psychological Association.

SHERIF, M. (1936). The psychology of social norms. Oxford, England: Harpter.

TRIPLETT, N. (1898). The dynamogenic factors in pacemaking and competition. American Journal of Psychology, 9, 507-533.

WEATHERLY, J. N., & BRANDT, A. E. (2004). Participant's sensitivity to percentage payback and credit value when playing a slot-machine simulation. Behavior and Social Issues, 13(1), 33-50.

WEATHERLY, J. N., DERENNE, A., & TERRELL, H. K. (2010). College students discount money "won" more than money "owed." The Psychological Record, 60, 463-472.

WEATHERLY, J. N., & MEIER, E. (2007). Studying gambling experimentally: The value of money. Analysis of Gambling Behavior, 1, 133-140.

WILSON, M., & DALY, M. (1985). Competitiveness, risk taking, and violence: The young male syndrome. Ethology & Sociobiology, 6, 59-73.

ZAJONC, R. B. (1965). Social facilitation. Science, 149, 269-274.

The authors thank Dean for serving as the dealer and Kyle Hill and Guy Keener for serving as confederates. Correspondence pertaining to this paper should be addressed to Casey L. McDougall, Acoma Behavioral Health Services, Indian Road 32, Acoma, NM, 87034 (e-mail: cmcdougall@puebloofacoma.org) or Jeffrey N. Weatherly, Department of Psychology, University of North Dakota, Grand Forks, ND 58202-8380 (e-mail: jeffrey.weatherly@und.edu).

Casey L. McDougall, Cheryl Terrance, and Jeffrey N. Weatherly University of North Dakota
In a few moments you will be given the opportunity to play a game
  of Blackjack. This game has been designed to function similarly to
  one that you would find in an actual casino. You will be staked with
  $5 in chips. Play as though you had just purchased them with your
  own money. There are four types of gaming chips worth different
  values. The red gaming chips are worth $.05; the white, $.10; the
  green, $.30; and the blue, $.50. You may wager a minimum bet
  of $.10 and a maximum bet of $.50, Blackjacks will pay 3 to 2. You
  may double down on any two cards and split any pair. However, you
  may only play at one player position at a time. You may end the
  session at any time. The session will end when A) you chose to quit
  playing, B) 20 minutes has elapsed, or C) you reach $.00 in gaming
  chips. You will be paid in cash at the end of the final session for
  the number of gaming chips you have accumulated over the course of
  the experiment. You may talk to the dealer during game play if you
  wish. Do you have any questions?


Table 1
Conditions Participants Experienced in the Six Sessions

Condition  Confederate present  Confederate wager  Confederate accuracy

Alone      No                                  --                    --
Stay       Yes                               $.50                  100%
Stay       Yes                               $.10                  100%
Stay       Yes                               $.50                   50%
Stay       Yes                               $.10                   50%
Leave      Yes                               $.30                   75%
Gale Copyright: Copyright 2011 Gale, Cengage Learning. All rights reserved.