Department of Psychology

Abstract:

The purpose of the research is to show that pool players use the process of chunking ball positions on the pool table. The experiments main objective was to see if there was a difference in chunking ball positions between novice and expert pool players. The participants were 10 student novice pool players and 10 student expert pool players and they consisted of both men and women. Students were displayed 9 visuals of ball positions on a pool table and were later asked to recall the ball positions by drawing what they remembered on a sheet of paper. Both novice and expert groups were given the same identical visuals. The results favored the expert chess players in their ability to recall more ball positions on the pool table than novice pool players. Although the results favored the experts over the novices it was very difficult to determine the underlying reasons behind their success in recalling more ball positions than novice pool players.

Introduction:

The research question is important because it will give a further analysis of the use of chunking information. It will show that chunking information develops over time for reasoning tasks. Since chess is very similar to pool, past studies and information on chess will be used to exploit the point that chunking information is used in pool. Experts in chess can recall a greater number of pieces on the board than novices and can recognize a greater number of chess patterns and each pattern contains more pieces than a pattern of a novice (Willingham, 2001).

It is predicted that an expert chess player has approximately 50,000 patterns in memory, a good player has 1,000, and a poor player has little but a few (Bedard & Chi, 1998). Highly skilled players are able to maintain a considerable amount of information in an accessible form during processing in domain-relevant tasks, for example, chess masters can play blindfolded (Ericsson & Delaney, 1998). A study regarding recall was performed with 26 participants and their recall of auditory presented chess positions. Results showed that skilled chess players were better at recalling typical game and random positions due to their ability to store more information in long-term memory (Saariluoma, 1989). Another study was done on expert and intermediate chess players attempting to pick the best move in five chess positions while monitoring their eye movements. Experts were faster and more accurate than intermediates when choosing the best move (Charness, Reingold, Pomplun & Stampe, 2001). One more study was with 18-49 year old chess players from novice to expert level and they were asked to recall up to 5 positions presented for 5 seconds. Percentage of recall with additional boards declined, but expert players recalled more pieces than novices (Gobet & Simon, 1996).

Chunking information in expert chess players and novice players is evident in a study where 2 novices studied a series of 500 chess positions over a certain period of time and their recall jumped from recalling 16% of them in the beginning to 40-50% after the session were done. This is strong evidence that chunking information develops over time with the consistency of playing a reasoning task, such as chess (Saariluoma & Laine, 2001). Expert chess players seem to show a superior ability to chunk relevant information.

Another study was a done using a computer where typical chess game positions and chess game random positions were presented on the screen varying from 1 second to 60 seconds. Afterwards the chess players were asked to recall both chess game and random positions. Researchers compared 21 chess players consisting of strong and weak chess players. The results demonstrated that strong players are better than the weak in recalling both game and random positions. These results were prevalent with long presentation times of chess positions, but were also true for short presentation times (Gobet & Simon, 2000). Expert players recall more chess positions because they hold a greater number of chess chunks of piece positions in long-term memory and are likely to associate a long-term chess chunk with any arrangement of chess positions. (Mcgregor & Howes, 2002). Consequently, highly skilled players are able to grab chunks from short-term memory, where, on the other hand, a novice player without the matching chunks can recall only a number of single pieces (Mcgregor & Howes, 2002). When its time to recall, the chess expert can unpack each chunk into its constituents to replace the pieces appropriately (Willingham, 2001).

The idea that experts can recall better than novices is demonstrated in a study done with Karate. Participants ranging from novice to expert level in Karate were shown a variety of karate techniques. After the karate techniques were displayed the participants had to recall the actual karate technique that was employed. Results showed that Karate novices could recall the same amount of karate techniques that were displayed as the Karate experts, but the Karate experts showed better recall in remembering the actual type of Karate techniques that were displayed (Bedon & Howard, 1992). Just like in chess and pool, karate has many positions and moves, which helps support the hypothesis that chunking is involved in pool.

As expert chess players have to reason and become aware of their piece positioning on the board, highly skilled pool players have to be alert of where their balls are on the table so that they can attack and defend efficiently. They always have to be thinking how to make it harder for their opponent to hit balls in the pocket and make it easier for themselves to hit balls in the pockets. Highly skilled players try to balance the attack and defense perspective of the game.

Similar to how chess players can acknowledge patterns of pieces on the board, pool players can do the same with patterns of positions of the balls on the table. Expert pool players should have more patterns of balls on the pool table in their mind than novice pool players.

Method:

The overall experiment is comparing the process of chunking in the game of pool between novice and expert chess players. As in previous studies researchers have found that expert chess players can recall more piece positions on the board compared to novice players. The experimenter used a very similar idea by using ball positions on the pool table. The participants had to recall how many ball positions are on the pool table after being presented a visual of ball positions on the pool table. The experimenter’s hypothesis is that expert pool players will recall more ball positions than novice pool players.

Participants:

The experiment was conducted on 20 Georgetown University students. The students were friends or acquaintances of the experimenter. The experimenter recruited the students at several social gatherings and also at the Billiards club near Wisconsin Ave in Georgetown. The students were asked if they could spare twenty minutes of their time to participate in a research project involving chunking memory and the game of pool. The participants were split up into two groups. One group consists of 10 novice pool participants and the second group consists of 10 expert pool participants. My participants were of both sexes and ranged from the freshman class to senior class. There were 3 female participants and 17 male participants. All three female participants were novice pool players. The novice participants were pool players who have never played on a steady basis and have played less than three times in the past six months. They have no firm acknowledgment of the game. The expert pool players had a long history of pool playing. They play on a consistent routine and have played pool at least twice a week for the past six months. This was confirmed by asking them face to face.

Apparatus:

The materials used in this experiment are: one room in a house, 9 visuals of ball positions on a pool table drawn on paper, paper, one watch, and one pen. Figure 2, 3, and 4 (see attached sheets) are examples of the visuals.

Procedure:

The test took place in a house, located at 1309 35th street NW in Georgetown. Before they began the experiment each participant had to sign an informed consent form. Each participant was advised that their information would be confidential and only used for research purposes. Afterwards each participant was presented visuals of ball positions on the pool table drawn on paper. Both novice and expert participant groups were shown the same identical visuals of ball positions on the pool table. They were given 9 visuals each ranging from three levels of difficulty. The first 3 visuals were level easy and they consisted of 5 ball positions on the pool table. The next 3 were level medium and they consisted of 7 ball positions on the pool table. The last 3 were level hard and they consisted of 9 ball positions on the pool table. Each visual was shown to the individual participant for 30 seconds. After the given time frame the participant had 30 seconds to recall how many ball positions they had in each given situation. There was a 30 second rest interval after recall was over before the next visual was displayed. The operational definition of chunking is the amount of ball positions recalled. An accurate ball position recall is the ability to remember the location of the ball on the pool table by drawing the ball’s location on an empty pool table drawn on a blank piece of paper during the recall period. The more ball positions recalled entails more chunking in memory. After the recall phase I asked each participant about what strategies they used when confronted with a visual of ball positions on the pool table and then asked to recall.

Design:

This is a between subjects quasi experimental design. The independent variables in the experiment will be the novice and expert pool participants and also the difficulty levels of the visuals of ball positions on the pool table. The dependent variables will be how many ball positions each participant can recall after the visual is removed from display.

Results:

Once the experimenter analyzed the drawings of recalled ball positions from each individual he properly scored each drawing (9 per person). When the experimenter summed up all the scores of the individuals in the novice and expert group the average score from each group was calculated. The results are shown in Figure 1A (see attached sheet) that the expert players on average recalled more ball positions on the pool table in all three difficulty levels. The data in Figure 1A were subjected to a 2 (expertise) × 3 (difficulty level) ANOVA shown in Figure 1B (see attached sheet). The results of the ANOVA indicate that there is a main effect of expertise level, F(1,174) = 31.86, p= <.0001 and a main effect of difficulty level, F(2,174) = 202.98, p = 5e – 7. Also there is an interaction between expertise level and difficulty level, F(2, 174) = 5.64, p = 0.004234. Evidence for the interaction is that the differences in amount of recalled ball positions between expert and novice level are not uniform through the easy, medium, and hard difficulty levels. There is not a significant difference between experts and novices at the easy level but there is one for the medium and hard difficulty level, thus the recall gap increases between novice and expert as the difficulty level of visuals gets harder.

Both the ANOVA and the graph in Figure 1C (see attached sheet) show that the experts faired a little better than the novices in recalling ball positions in the easy, medium, and hard difficulty level. The difference in experts scores compared to novices increased from the easy to hard difficulty level. The error bars on the graph for novice and expert did however overlap for the easy difficult level, but not for the medium and hard difficulty level. This indicates that the mean for the experts could be the same as novices in the easy level, but not for the medium and hard difficulty level. These findings infer that there is an interaction between difficulty level and participant skill level.

Discussion:

The results that were obtained in the experiment demonstrate that expert pool players compared to novice pool players can recall more ball positions under the medium and hard difficulty level and they recall about the same for the easy difficulty level. The experimenter’s alternate hypothesis that expert pool players will recall more ball positions than novice pool players is supported by the concluding results and also indicates that the null hypothesis is rejected. These results mean that expert pool players already have information for certain pool ball positions instilled in their memory. Particular ball positions may seem familiar to an expert pool player and these positions are embedded in the player’s memory in chunks. An expert pool player will have certain ball positions in memory chunks implanted in their head that will enhance their recall performance.

The strategies that the participants reported to the experimenter included chunking the ball positions that were in the same vicinity on the pool table together first and than trying to remember where the other balls on the pool table were and the other strategy was the opposite where the participants reported chunking the balls that were not close together first and than trying to remember the balls that were in the same location on the pool table. All of the participants stated that chunking the ball positions was not necessary for the easy difficulty level, but was for the medium and hard difficulty level. This particular comment supports the results of the study where both novice and expert participants faired about the same for the easy difficulty level, but not for the other two difficulty levels because chunking was not used for the easy difficulty level.

In this experiment the experimenter controlled a lot of variables. The experimenter controlled the setting and the participants that were used for both novice and expert groups. Further the experimenter manipulated the display of visuals presented to the participant. Since everything was done in one sitting and in the same setting the experimenter controlled the amount of distractions the participant will endure. By performing this method the experimenter eliminated any distractions that might occur in a natural environment. On the same note there is a bias at hand because the participants knew what the experiment was about and the experimenter gave them certain instructions for the experiment that may have altered or interfered with a natural response.

Another limitation to this experiment is that the experimenter only used one type of chunking. The experiment only dealt with the knowledge of the positioning of balls on the pool table. This is only one type of encoding memory.

The theoretical implications of this particular experiment are that it further supports the claim that reasoning tasks similar to chess use the process of chunking. Related to how expert chess players can recall more positions this experiment has shown that expert pool players can recall more ball positions than novices, which gives a stronger link between the game of chess and the game of pool.

A future study that can be conducted in regards to chunking information in memory can involve choosing the best move out of five options. For example, given a ball situation, a participant is to select which balls to hit in the pocket that will enhance his or her attack, while diminishing their opponents attack at the same time. The pool participants may fair better or worst in other chunking experiments, but undoubtedly this experiment has given more evidence towards the use of chunking information in memory in the reasoning tasks, such as the game of pool.

References:

Bedard, J.,& Chi, T.H.(1998). Expertise. Introductory readings for Cognitive Psychology, 3, 178-179.

Bedon, B.G., & Howard, D.V. (1992). Memory for the frequency of occurrence of karate techniques: A comparison of experts and novices. Bulletin Of the Psychonomic Society, 30, 117-119.

Charness, N., Reingold, E.M, Pomplun, M., & Stampe, D.M. (2001). The perceptual aspect of skilled performance in chess: Evidence from eye movements. Memory and Cognition, 29, 1146-1152.

Ericsson, A.K., & Delaney, P.F. (1998). Working Memory and Expert Performance. Working Memory and Thinking, 94.

Gobet, F., & Simon, H.A. (1996). Templates in chess memory: A mechanism for recalling several boards. Cognitive Psychology, 31, 1-40.

Gobet, F., & Simon, H.A. (2000). Five seconds or sixty? Presentation time in expert memory, 24, 651-682.

McGregor, S.J., & Howes, A. (2002). The Role of attack and defense semantics in skilled player’s memory for chess positions. Memory and Cognition, 30, 707-717.

Saariluoma, P., & Laine, T. (2001). Novice Construction of chess memory. Scandanavian Journal of Psychology, 42, 137-146.

Saariluoma, P. (1989). Chess players recall of auditory presented chess positions. European Journal of Cognitive Psychology, 1, 309-320.

Willingham, D.B. (2001). Cognition. The Thinking Animal. Saddle River, New Jersey: Prentice Hall.