Printer Friendly

Effects of anagrams with single or multiple solutions, level-of-processing, and generation on implicit memory.

In two experiments I investigated the nature of implicit memory test, anagram solution, by manipulating several variables during the study phase. In Experiment 1, the effects of solution type (single vs. multiple), and study type (physical orienting vs. semantic orienting vs. control) on an anagram solution test were examined. Results of an ANOVA revealed significant main effects of solution type manipulation (single vs. multiple). In Experiment 2, the effects of solution type (single vs. multiple), and study type (reading vs. generate vs. control) on the anagram solution test were examined. ANOVA revealed significant main effects of solution type and generation manipulation. In all study types the anagrams with single solution produced more priming than anagrams with multiple solutions. The findings showed that the anagram solution test was an implicit test that had a perceptual nature and the number of anagram solutions was important for the test performance. Dissociations were obtained between test materials (single vs. multiple) used to measure the same memory processes. I emphasized dissociations among constructional features (solution type) of an anagram solution test.

Memory has been defined traditionally as the faculty of retaining and recalling past experiences. However, today the existence of implicit memory has been shown in some experimental studies (Challis, 1989; Eysenck, 1990).

Even though different researchers name it differently, the most used term is implicit memory, suggested by Graf and Schacter (1985). However, as mentioned by Schacter (1987) in his review article on this issue, the history of implicit memory goes back earlier, even though it became popular for the last twenty years. Today, implicit memory that can be defined as the recollection of knowledge unconsciously is different than explicit memory in which knowledge is recalled consciously. In traditional explicit memory tests (recognition, recall), a target personal event of the subjects related to past experience is attributed intentionally.

However, in implicit memory tests there is not any attribution to the past experiences. Tests like word stem completion, word fragment completion, perceptual identification, and anagram solution are the most used tests in the assessment of implicit memory. In these tests, the inclination of the subjects to complete the test items with the words that they have faced in the study phase, namely priming, is accepted as the sign of an implicit memory (Greene, 1992; Richardson-Klavehn & Bjork, 1988; Roediger, 1990; Schacter, 1987; Schacter, Chiu & Ochesner, 1993). Interest in implicit memory tests has increased as a result of several studies that have shown dissociations between performance on implicit and traditional explicit memory tests (Bowers & Schacter, 1990; Graf & Mandler, 1984; McCauley, Eskes & Moscovitch, 1996; Paller, 1990; Schacter & Garf, 1989).

When the related literature is examined, it can be seen that implicit memory tests can be categorized under two main titles: Perceptual and conceptual tests. In perceptual implicit memory tests (data-driven), the subjects are expected to react to the test materials that have a perceptual nature; that is to say, the subjects are asked to find solutions by assessing the stimulus on a perceptual-physical base. In conceptual memory tests (conceptually-driven), the subjects are expected to react to the test materials that have a conceptual nature. In other words, they are supposed to find solutions by assessing the stimulus on a conceptual-semantic base (Blaxton, 1989; Roediger & Blaxton, 1987; Roediger, Srinivas & Weldon, 1989; Tulving & Schacter, 1990). There are also many studies that call attention to the fact that implicit memory tests may vary even in themselves (McDermott & Roediger, 1994; McDermott & Roediger, 1996; Rajaram & Roediger, 1993; Schacter, 1984; Srinivas & Roediger, 1990; Weldon, 1991).

When the studies on implicit memory tests are examined, it can be seen that word stem completion (Crabb & Dark, 1999; McCauley, Eskes & Moscovitch, 1996), word fragment completion (McDermott & Roediger, 1994; Rajaram & Roediger, 1993; Weldon, 1991), and perceptual identification (Rajaram & Roediger, 1993; Weldon, 1991) are widely used in perceptual implicit memory tests.

Category verification (Light, Prull & Kennison, 2000), category exemplar production (Light, Prull & Kennison, 2000; Mulligan, Guyer & Beland, 1992), and answering general knowledge (Blaxton, 1989; Hamann, 1990) take place in the group of conceptual implicit memory tests.

Roediger, Weldon and Challis (1989) proposed to classify tests along the continuum between data-driven and conceptually-driven processing using three criteria. (1) Level-of-processing (LOP) effect: LOP has always affected explicit memory tests and conceptual implicit memory tests but not perceptual implicit memory tests. In other words, the performance following the semantic orienting tasks in conceptual implicit tests is higher than the performance in physical orienting tasks. However, performance in perceptual implicit memory tests is not affected by this manipulation (Graf & Mandler, 1984; Hamman, 1990; Jacoby & Dallas, 1981; Thapar & Greene, 1994). (2) Generation effect: If performance on a test is better in the read condition than in the generate condition it is called a 'perceptual implicit memory test' and if it is vice-versa it is called a 'conceptually implicit memory test' (Jacoby, 1983; Slameka & Graf, 1978, Srinivas & Roediger, 1990). (3) Modality effect: The change of modality in study-test phase has often affected perceptual implicit memory tests but not the conceptually implicit and explicit memory tests. It is a known fact that the presentation modality used in the study-test phase affects the performance of perceptual implicit tests. Because the change in the modality affects the perception of data this affects the performance on the tests that are sensitive to perceptual features (Srinivas & Roediger, 1990). Furthermore, the visual presentation of the words has a stronger priming effect on perceptual implicit memory tests and their auditory presentation has a stronger priming effect on explicit and conceptual implicit memory tests (Weldon, 1991; Weldon & Coyote, 1996; Weldon & Roediger, 1987).

These findings have been explained by the transfer-appropriate processing (TAP) approach. The TAP account of dissociations among memory tests led us to predict that the tests would show parallel effects; this prediction was based upon the theory's assumption that conceptual tests will behave similarly as a function of various independent variables (McDermott & Roediger, 1996). According to this approach, study-test transfer depends on the overlap between the types of processing at study and test phase. This approach generally argues that the effect of a prior experience on a subsequent test depends on the match between the mental processes engaged by the study experience and the mental processes required by the test. To extend the mental processes required at study overlap with those at test, test performance will be improved (Roediger, Weldon & Challis, 1989). Mental processes are very specific. Reading the word CAT requires different mental processes than reading the word DOG. Reading the word DOG involves different mental processes than hearing the word DOG. Thus the modality effect is explained as a mismatch of processes.

On the other hand, it is still debatable as to which classification (perceptual or conceptual) some of the implicit memory tests like anagram solution should be included. Anagram solution scores correlate positively with word identification test scores. However, anagram solution scores are not correlated with implicit memory test scores such as perceptual clarification, and word fragment completion or with explicit memory test scores such as recall and recognition (Perruchet & Baveaux, 1989). The anagram solution test is similar to the word fragment completion test, that is to say, it is mostly a data-driven test (Jacoby & Dallas, 1981). According to the results of a series of experiments conducted by Srinivas and Roediger (1990), performance on an anagram solution test is not affected by manipulation of generation. In other words, where there is a reading (no context) condition, anagram solution test performance is higher than the generation condition (reading > generation), but the difference is not significant. On the other hand, it is affected by presentation modality (visual > auditory) and it shows a perceptual-driven property. Another contradictory point is that, modality effect can be observed in a semantic oriented task whereas physical oriented task cannot be observed. In the light of those findings, even though it is not very clear-cut, the anagram solution is largely a perceptual-driven implicit memory test. However, it also has some conceptual-driven properties. In other words, the anagram solution test seems much closer to the conceptual processing pole on the perceptual-conceptual processing line when compared to other implicit memory tests that have a perceptual nature (Srinivas & Roediger, 1990).

In a study in which word stem completion, word fragment completion, word identification and anagram solution tests are compared, it has been observed that the first three tests are very similar to each other, whereas anagram solution tests are less similar to them. Other research has emphasized that even though the anagram solution test is largely a perceptual test, further research may be needed to clarify this matter (Rajaram & Roediger, 1993; Srinivas & Roediger, 1990).

The anagram solution is a test that is based on forming words from letters that are chosen at random. The argument on the nature of the test has been focused on the construction of the anagrams. As a matter of fact, in some of the studies the row of the letters in words are chosen randomly, and in other studies the first or last three letters of a word are left unchanged (Jacoby & Dallas, 1981). In this way, the test has gained a perceptual property. However, McAndrews and Moscovitch (1990) argued that to fix the first letters of words does not cause a significant difference in performance.

The first aim of the present study is to determine if the anagram solution test score is affected by the manipulation of LOP and generation. In general, the performance in implicit memory tests is independent of subject strategies like level-of-processing (Jacoby & Dallas, 1981; Roediger, Weldon & Challis, 1989; Thapar & Greene, 1994), paying attention/ignoring (Crabb & Dark, 1999), and forget/recall instructions (Paller, 1990). However, when the literature is examined, it can be seen that, contrary to other implicit memory tests, the number of the studies that test the effects of LOP and generation on anagram solution test performance is limited (Jacoby & Dallas, 1981; Srinivas & Roediger, 1990).

The second aim of the present study is to determine whether the anagram solution test score is affected by the solution type manipulation (anagram's being with single or multiple solution).

In sum, the effects of the type of the solution (single solution vs. multiple solution), the LOP (physical orienting vs. semantic orienting vs. control) and the generation (read vs. generate vs. control) on the anagram solution test performance was examined in this study.

EXPERIMENT 1

Method

Participants Sixty-six undergraduate female students who took an 'Introduction to Psychology' course at Hacettepe University - Turkey participated in this experiment. Their ages ranged from 18 to 23 years (M= 19.7). Undergraduates participated in the experiment for extra course credit. All of the participants were volunteers, native speakers of Turkish and reported normal or corrected-normal vision and audition.

Materials For both lists, single and multiple solution, a total of 34 Turkish words were selected (four of those were buffer words). Parameters for word selection were Cangoz (1997) to be high to moderate frequencies (university students sample in Turkey), concrete objects, each of the anagrams in the multiple solution list had at least two solutions, and in the single solution list had only one solution in Turkish. There were 30 words in the database that fit these parameters. During the study phase, the participants were presented cards on which the words were written separately. During the test phase, the participants were given an anagram solution test that consisted of 30 anagrams. Randomly dispersed letters formed the test. The participants were asked to form a word by using those dispersed letters. Fifteen of those anagrams had a single solution (NRA, LOABN, BRDAAK, HRFA, IKED, ALTAC, RIFAC, EBIBR, TIBLE, CIKLE, VYAA, ZAAKK, EEPT, KCAAL, OZG) and 15 had multiple solutions (MAAS, KNBAA, RKAI, CLKAA, RATKA, VATNA, TKEE, ILF, KCIEL, AMC, TPLAO, TKIIL, RENIH, TIAK, UUKT). The anagrams with single solution had only one possible solution in Turkish (e.g. LOABN; solution: BALON). The anagrams with multiple solutions had at least two solutions in Turkish (e.g., KNBAA; solutions: BANKA, KABAN, BAKAN). The anagrams with multiple solutions had an average of three solutions. Anagrams with single and multiple solutions were matched according to the number of the letters (the length of the words with single solution = with multiple solution = M= 4.6, range= 3-6) and number of the syllables (with single solution = with multiple solution= M= 4.4, range= 3-5). The words presented in the study phase were formed by the one possible solution of the anagrams that had a single solution and one of the solutions of the anagrams with multiple solutions.

Design A 3 (study type: physical orienting, semantic orienting, control) x 2 (solution type: with single solution, with multiple solution) mixed design was used. Solution type conditions were manipulated within subjects. Items were divided into two sets of 15 items each. Each participant studied these two sets (single solution, multiple solution). Study type conditions were manipulated between sixty-six subjects participating in each of the three study conditions (physical orienting, semantic orienting, and control). The dependent variable was the proportion of the correctly solved responses in the anagram solution test (responses given as the target word). In other words, the number of answers which contained target words that the participants faced in the study phase, was used as the measure of implicit memory performance (priming effect).

Procedure Participants were tested individually. They were designated randomly to each three study conditions. Participants in the physical orienting condition were required to find the number of vowels in the words whereas the participants in the semantic orienting condition were asked to rate on a five point scale how much they liked or disliked each word. The participants in the control group were not subjected to any of the study conditions.

The words were presented randomly on cards of 21 x 15 cm. written with black pen all in capital letters. Each word was presented for 6 seconds. In order to prevent probable effects of primacy and recency, two buffer words were added to the beginning and ending of the words, but the performance related to those words was not added to the assessment. After the study phase, all the participants were given a 5 minute distractor task. In the distractor task the participants were asked to write down the names of as many Turkish authors as possible within the 5 minute period of time. The participants that were assigned to each of these three study conditions were presented an anagram solution test in which the anagrams with single and multiple solutions were presented randomly. In other words, all participants solved both the anagrams with single solutions and anagrams with multiple solutions within the scope of the same anagram solution test. Similar to the study phase, the anagrams were written on the cards of 21 x 15 cm. with black pen and all in capital letters. Participants were asked to write down the first word that came to their mind by using the mixed and puzzle letters. The answers participants gave were recorded on answer sheets by the experimenter.

Results

The proportion of responses correctly solved (responses given as the target word) as a function of the study (physical orienting, semantic orienting, control) and solution type (single, multiple) conditions are presented in Table 1. In the single solution condition the results showed priming effects for semantic orienting (19%) and physical orienting tasks (26%) as compared to the control type, with a 7% advantage of physical orienting over the semantic orienting type. Results indicated priming effects for semantic orienting (20%) and physical orienting tasks (29%) as compared to the control type, with a 9% advantage of physical orienting as compared to the semantic orienting task under anagrams in the multiple solution type.

A 3 (study type: physical, semantic, control) x 2 (number of solutions: single solution, multiple solution) mixed analysis of variance (ANOVA), with study type condition serving as a between-subjects variable and solution type condition as a within-subjects variable, was performed on the data. The results of these analysis indicated a reliable main effect of solution type condition, F (1, 63)= 104.85, MSe= .005, p< .01, but the study condition and the interaction between the two was not significant. According to the results of the ANOVA for solution type variable, the score of the anagrams with single solution test is higher than anagrams with multiple solutions under any study condition. Tukey Post-Hoc tests were used for comparisons between means. The critical comparison between the physical orienting and semantic orienting task conditions was not significant. In addition, the difference between physical and semantic orienting task condition and control condition was significant (respectively, [t (3, 63)= 26.13, p < .01; t (3,63)= 18.92, p < .01]. This finding indicates the existence of a significant priming effect under physical and semantic orienting conditions.

Discussion

The results of Experiment 1 show that the anagram solution test was not affected by LOP manipulation, both in single and multiple solution. Perceptual implicit memory tests should not be affected by LOP manipulation (Roediger, Weldon & Challis, 1989; Roediger, Weldon, Stadler & Riegler, 1992). In addition, anagrams with single solutions yielded higher performance than the anagrams with multiple solution. Thus, anagram solution test performance was affected by solution type.

EXPERIMENT 2

Method

Participants Ninety undergraduate students who took an 'Introduction to Psychology' course at Hacettepe University-Turkey participated in this experiment (45 males and 45 females). Their ages ranged from 18 to 26 years (males M= 20.1, females M= 18.7). Undergraduates participated in the experiment for extra course credit. All of the participants were volunteers, native speakers of Turkish and reported normal or corrected-normal vision and audition. The number of male and female participants was equal for all the experimental conditions.

Materials The verbal materials for this experiment were the same as those of Experiment 1. During the test phase the participants were given the same anagram test used in the first experiment. In the generate condition sentence frames were used to guide the generation of 30 target words. The sentence contexts were selected so that they specified only the target word. These materials were normed with 40 participants in order to make sure that the correct words were generated 95% of the time. In the read condition participants were instructed to read aloud 30 target words which had no context.

Design A 3 (study type: generate, read, control) x 2 (solution type: with single solution, with multiple solution) mixed design was used. Solution type conditions were manipulated within subjects. Items were divided into two sets of 15 items each. Each participant studied these two sets (single solution, multiple solution). Study type conditions were manipulated between ninety subjects participating in each of the three study conditions (generate, read, control). The dependent variable was the number of correct answers given in the anagram solution test (responses given as the target word). In other words, the number of answers containing target words that the participants faced in the study phase, was used as the measure of implicit memory performance (priming effect).

Procedure Participants were tested individually and were assigned to one of three study conditions randomly. The participants within the generate condition were asked to complete the sentences with the target words whereas the participants within the read condition were asked to read the target words without a context. The participants within the control group were not subjected to any of the study conditions.

In the generate condition, a form which consisted of 30 incomplete sentences listed randomly was used. In the read condition, the words were presented randomly on the cards of 21 x 15 cm, written with black pen and all in capital letters. Each word was presented for 6 seconds. In order to prevent probable effects of primacy and recency, two buffer words were added to the beginning and ending of the words, but the performance related to those words was not added to the assessment. After the study phase, all the participants were given a 5 minute distractor task. The procedure for the distractor task phase and test phase of Experiment 2 was the same that of Experiment 1. As in Experiment 1, the participants that were assigned to each of the three study conditions were presented an anagram solution test in which the anagrams with single and multiple solutions were presented randomly.

Results

The proportion of responses correctly solved (responses given as the target word) as a function of study type (read, generate, control) and solution type (single, multiple) conditions were presented in Table 2. The results showed priming effects for the generate (28%) and read tasks (35%) with a 7% advantage of read as compared to the generate task under the anagrams with single solution condition. And data indicated priming effects for the generate (16%) and read tasks (27%) with an 11% advantage of read to the generate task under the anagrams with multiple solution condition.

A 3 (study type: generate, read, control) x 2 (number of solutions: single solution, multiple solution) mixed analysis of variance (ANOVA), with study type condition serving as a between-subjects variable and solution type condition serving as a within-subjects variable, was performed on the data. The results of these analysis indicated a reliable main effect of solution type condition, F (1, 87) = 313.53, MSe= 2.054, p < .01, and study condition, F (2, 87)= 92.23, MSe= 1.525, p< .01, finally the interaction between two was significant F (2, 87)= 9.23, Mse = .06, p<. 01. That was, the main effect of the solution and study type variables and the interaction of those two were significant. Tukey Post-Hoc tests were used for comparisons between means. The critical comparison between the read, generate and control conditions was significant [t (3, 87) = 9.62, p < .01]. That is, the score of anagram solution test under the read condition was significantly higher than the generate condition. In addition, the difference between read, generate and control conditions was significant (respectively, [t (3, 87) = 29.8, p < .01; t (3,87)= 20.2, p < .01]. These findings showed that there was a priming effect under the read and generate task conditions.

According to Tukey results related to interaction effect that was significant, the priming effect decreases in conformity with read, generate and control conditions for the anagrams with single solution (respectively, t (2, 87)=6.58, p < .01; t (2,87)= 32.03, p < .01, t (2,87)= 25.45, p < .01]. For the anagrams with multiple solution, the priming effect decreases in conformity with read, generate and control conditions (respectively, t (2, 87) =10.34, p < .01; t (2,87)= 24.45, p < .01, t (2,87)= 14.12, p < .01]. In summary, under any study condition, the anagrams with single solution had higher scores than the anagrams with multiple solutions. In addition, for both solution types, the score for the anagram solution test under the read condition was higher than the score for the generate condition.

Discussion

The results of Experiment 2 show that the anagram solution test, which is an implicit memory test, was affected by read-generate manipulation, both in single and multiple solutions, and this effect was positively higher for the reading task.

An implicit memory test with a perceptual nature was expected to be affected by read-generate manipulation and this effect was expected to be in favor of reading (Roediger, Weldon & Challis, 1989; Roediger, Weldon, Stadler & Riegler, 1992). In this context, the findings of Experiment 2 indicate that an anagram solution test was an implicit test with a perceptual nature. Even though these findings supported the findings of Srinivas and Roediger's (1990) in their research, the difference between read and generate conditions was not statistically significant; but in our study, this difference was significant. Anagrams with a single solution yielded a higher performance than the anagrams with multiple solutions for both the read and generate tasks. However, this variable did not have a crucial role in determining the nature of the test.

GENERAL DISCUSSION

The primary purpose of the present experiments was to classify the anagram solution test by observing how they responded to manipulation of variables that have well-documented effects (LOP, generation) on the other implicit memory tests. That is, if a task or test is not affected by LOP manipulation and if the read task yields higher scores than the generate task and if it is sensitive to modality manipulation it is called a data-driven or perceptual implicit test (Roediger, Weldon & Challis, 1989).

According to TAP, study-test transfer depends on the overlap between the types of processing at study and test phase. This approach generally argues that the effect of a prior experience on a subsequent test depends on the match between the mental processes engaged by the study experience and the mental processes required by the test. When the mental processes required at study overlap with those at test, test performance will be improved (Roediger, Weldon & Challis, 1989). Reading a word requires different mental processes than generating a word. If this study phase (reading/physical coding) was matched with the test phase (anagram solution test), implicit memory performance was higher.

In this context, according to the findings of Experiment 1, anagram solution test has not been affected by LOP manipulation. The test performance in a physical orienting task is equal to a semantic orienting task. These findings were not supported by the TAP. According to the findings of Experiment 2, the anagram solution test was affected by the read-generate manipulation. The test performance on the read task was found to be significantly higher than on the generate task. The finding shows that the test has a perceptual nature. These findings supported the TAP.

The present study also aims at testing the effects of the number of solutions that may cause ambiguities about classifying the anagram solution test in the research literature. I doubt that weakly data-driven effect (not significant) obtained from different studies that examine the nature of the anagram solution test might result from the construction of the anagrams used (Jacoby & Dallas, 1981; McAndrews & Moscovitch, 1990; Rajaram & Roediger, 1993; Srinivas & Roediger, 1990). As a matter of fact, in some of the studies this test used, the row of the letters was determined randomly (Rajaram & Roediger, 1993; Srinivas & Roediger, 1990), in others, the first or last three letters of the words were kept fixed (Jacoby & Dallas, 1981). However, McAndrews and Moscovitch (1990), in a study they conducted by using anagrams formed by different rules, showed that to keep the first letters of the words fixed does not cause any significant difference.

In the present study, the number of solutions that forms the anagram has been manipulated and its effects on test performance have been examined. The effect of this variable has not been examined in other studies. The main effect of solution type was found significant both in Experiment 1 and 2. In addition to this, the interaction of study and solution type in Experiment 2 was also significant. In other words, under all study conditions (physical orienting vs. semantic orienting and read vs. generate) anagrams with single solution yielded higher performance than anagrams with multiple solutions.

All these results point out the importance of the constructional features of the items (anagrams) that the anagram solution test comprises. Anagrams with single solutions have only one solution in Turkish and the subjects encountered those solutions. Yet, anagrams with multiple solutions had at least two solutions and the subjects encountered only one of those solutions. That is to say, the probable answer for the anagram might be different than the word presented in the study phase. Consequently, when the anagrams with a single solution were used, the participants may have solved it either because they knew the word from the study phase (priming effect) or because it was the only probable answer. Thus, the use of anagrams with a single solution helped the test gain a more perceptual property.

Within this context, according to the results of these present experiments, the number of the solutions for the anagrams should be taken into consideration, and if possible, should be chosen among the anagrams with single solutions. To use the anagrams with single and multiple solutions together or in different numbers might cause the number of solutions for anagrams to play a confounded variable role. It was possible to explain the results of these experiments with TAP. According to this approach, study-test transfer depends on the overlap between the types of processing at study and test phase. This approach generally argues that the effect of a prior experience on a subsequent test depends on the match between the mental processes engaged by the study experience and the mental processes required by the test. When the mental processes required at study overlap with those at test, test performance will be improved (Roediger, Weldon & Challis, 1989). This approach contends that the performance will increase depending on the similarities between the mental process that is needed for the encoding during the study phase and the mental process that is needed for the test. Within this context, perceptual implicit memory tests had a nature that necessitate the subjects find solutions by making perceptual assessments and by focusing on the physical features of the stimulus. The performance in those tests depends on perceptual system (Blaxton, 1989; Roediger & Blaxton, 1987; Schacter, 1994). The findings of the present study showed that LOP type was not affected by anagram solution performance. But, physical orienting coding was higher than semantic oriented coding in anagram solutions test. In other words, following the study phase (to find the number of the vowels in the target word or to read the word) that necessitates the subjects to focus on physical features of the stimulus for a perceptual assessment, to test them in a task where they form words by using the mixed letters has increased the priming in the test. As mentioned before, even though in some of the studies (Srinivas & Roediger, 1990; Rajaram & Roediger, 1993) it was claimed that anagram solution is a perceptual test whose conceptual components are very rich or that the findings related to anagram solution were unclear and vague, by this study some clear-cut findings have been put forward that anagram solution test was a perceptual implicit memory test. The reason for this might be that, in other studies the effect of solution type has not been taken into consideration. Likewise, in general, it is accepted that performance in memory tests, depending on the requirements of the test, is affected by level-of-processing in different ways (Kolers and Roediger, 1984; Weldon, 1991). The findings of these experiments were important since they show that the test material in some of the implicit memory tests may be affected.

In future studies other implicit memory tests with multiple solutions (word fragment completion, word identification) can also be examined in order to see whether they are affected by the number of solutions. These studies will contribute to the classification and better understanding of the nature of implicit memory tests.

The results of two experiments summarized above support five points: a) The score of the anagram solution test was affected by reading condition. Under read-generate manipulation, read task will cause more priming than generate task. b) The score on the anagram solution test was not affected by level-of-processing. c) According to (a) and (b), anagram solution was an implicit memory test that had a perceptual nature. d) The score of the anagram solution test was affected by the number of solutions. Under this manipulation, single solution anagrams will cause more priming than multiple solution anagrams. e) Results of Experiment 1 did not support TAP approach, but results of Experiment 2 did support the TAP approach.

Appendix A

Study Materials in Experiment 1 and 2

NAR, BALON, BARDAK, HARF, KEDI, CATAL, FIRCA, BIBER, BILET, CILEK, AYVA, KAZAK, TEPE, CAKAL, GOZ, MASA, KABAN, RAKI, CAKAL, TARAK, TAVAN, ETEK, FIL, CILEK, CAM, PALTO, NEHIR, ATKI, KUTU
Appendix B
Test Materials (Anagrams) in Experiment 1 and 2

Anagrams     Single solution
NRA              NAR
LOABN            BALON
BRDAAK           BARDAK
HRFA             HARF
IKED             KEDI
ALTAC            CATAL
RIFAC            FIRCA
EBIBR            BIBER
TIBLE            BILET
CIKLE            CILEK
VYAA             AYVA
ZAAKK            KAZAK
EEPT             TEPE
KCAAL            CAKAL
OZG              GOZ

Anagrams     Multiple solution
MAAS             MASA, ASMA
KNBAA            KABAN, BAKAN, BANKA
RKAI             RAKI, IRAK, KARI
CLKAA            CAKAL, KALCA, ALCAK
RATKA            TARAK, KATAR
VATNA            TAVAN, VATAN
TKEE             ETEK, TEKE
ILF              FIL, LIF
KCIEL            CILEK, CELIK
AMC              CAM, MAC
TPLAO            PALTO, TOPAL, POLAT
TKIIL            KILIT, TILKI
RENIH            NEHIR, REHIN
TIAK             ATKI, TAKI
UUKT             KUTU, UTKU


REFERENCES

Blaxton, T.A. (1989). Investigating dissociations among memory measures: Support for a transfer-appropriate processing framework. Journal of Experimental Psychology: Learning, Memory & Cognition, 15, 657-668.

Bowers, J.S. & Schacter, D.L. (1990). Implicit memory and test awareness. Journal of Experimental Psychology: Learning, Memory & Cognition, 16, 404-416.

Cangoz, B. (1997). Calisma kosulu turunun farkli bellek olcumleri uzerindeki etkileri [Effects of study condition type on different memory measurements]. Psikiyatri, Psikoloji ve Psikofarmakoloji Dergisi [Journal of Psychiatry, Psychology and Psychopharmacology], 7(2), 106-115.

Challis, B.H. (1996). Implicit memory research in 1996: Introductory remarks. Canadian Journal of Experimental Psychology, 50 (1), 1-4.

Crabb, B.T., & Dark, V. (1999). Perceptual implicit memory requires attentional encoding. Memory & Cognition, 27(2), 267-275.

Eysenck, M.W.(Ed.).(1990). The Blackwell dictionary of cognitive psychology. Oxford: Basil Blackwell.

Graf, P. ,& Mandler, G. (1984). Activation makes words more accessible, but not necessarily more retrievable. Journal of Verbal Learning and Verbal Behavior, 23, 553-568.

Graf, P. ,& Schacter, D.L. (1985). Implicit and explicit memory for new associations in normal and amnesic subjects. Journal of Experimental Psychology: Learning, Memory & Cognition, 11, 501-518.

Graf, P., Squire, L.R. & Mandler, G. (1984). The information that amnesic patients do not forget. Journal of Experimental Psychology: Learning, Memory & Cognition, 10 (1), 164-178.

Greene, R.L. (1992). Human memory: Paradigms and paradoxes. New Jersey: LEA.

Hamman, S.B. (1990). Level-of-processing effects in conceptually-driven implicit tasks. Journal of Experimental Psychology: Learning, Memory and Cognition, 16, 970-977.

Jacoby, L.L. (1983). Remembering the data: Analyzing interactive processes in reading. Journal of Verbal Learning and Verbal Behavior, 22, 485-508.

Jacoby, L.L., & Dallas, M. (1981). On the relationship between autobiographical memory and perceptual learning. Journal of Experimental Psychology: General, 110, 306-340.

Kolers, P.A. & Roediger, H.L. (1984). Procedures of mind. Journal of Verbal Learning and Verbal Behavior, 23, 425-449.

Light, L.L., Prull, M.W.,& Kennison, R.F. (2000). Divided attention, aging, and priming in examplar generation and category verification. Memory & Cognition, 28(5), 856-872.

McAndrews, M.P., & Moscowitch, M. (1990). Transfer effects in implicit memory. Journal of Experimental Psychology: Learning, Memory and Cognition, 16, 772-788.

McCauley, M.E., Eskes, G. & Moscovitch, M. (1996). The effect of imagery on explicit and implicit tests of memory in young and old people: A double dissociation. Canadian Journal of Experimental Psychology, 50(1), 34-40.

McDermott, K.B., & Roediger, H.L. (1994). Effects of imagery on perceptual implicit memory tests. Journal of Experimental Psychology: Learning, Memory and Cognition, 20(6), 1379-1390.

McDermott, K.B., & Roediger, H.L. (1996). Exact and conceptual repetition dissociate conceptual memory tests: Problems for transfer appropriate processing theory. Canadian Journal of Experimental Psychology, 50(1), 57-71.

Mulligan, N.W., Guyer, P.S., & Beland, A. (1999). The effects of levels-of-processing and organization on conceptual implicit memory in the category examplar production test. Memory & Cognition, 27(4), 633-647.

Paller, K.A. (1990). Recall and stem completion priming have different electrophysiological correlates and are modified differentially by directed forgetting. Journal of Experimental Psychology: Learning, Memory and Cognition, 16, 1021-1032.

Perruchet, P., & Beveaux, P. (1989). Correlational analysis of explicit and implicit memory performance. Memory & Cognition, 17, 77-86.

Richardson-Klavehn, A., & Bjork, R.A. (1988). Measures of memory. Annual Review of Psychology, 39, 475-543.

Rajaram, S., & Roediger, H.L. (1993). Direct comparison of four implicit memory tests. Journal of Experimental Psychology: Learning, Memory, and Cognition, 19(4), 765-776.

Roediger, H.L. (1990). Implicit memory: Retention without remembering. American Psychologist, 45, 1043-1056.

Roediger, H.L., & Blaxton, T. (1987). Effects of varying modality, surface futures, and retention interval on priming in word-fragment completion. Memory & Cognition, 15, 379-388.

Roediger, H.L., Srinivas, K. & Weldon, M.S. (1989). Dissociations between implicit measures of retention. In S. Lewandowsky, & K. Kirsner (Eds.), Implicit memory: Theoretical issues. (pp. 67-87). Hillsdale, NJ: Erlbaum.

Roediger, H.L., Weldon, M.S., & Challis, B.A. (1989). Explaining associations between implicit and explicit measures of retention: A processing account. In H.L.Roediger & F.I.M. Craik (Eds.), Varieties of memory and consciousness: Essays in honour of Endel Tulving (pp. 3-41). Hillsdale: Erlbaum.

Roediger, H.L., Weldon, M.S., Stadler, M.L., & Riegler, G.L. (1992). Direct comparison of two implicit memory tests: Word fragment and word stem completion. Journal of Experimental Psychology: Learning, Memory, and Cognition, 18, 1251-1269.

Schacter, D.L. (1987). Implicit memory: History and current status. Journal of Experimental Psychology: Learning, Memory, and Cognition, 13, 501-518.

Schacter, D.L. (1994). Priming and multiple memory systems: Perceptual mechanisms of implicit memory. In D.L. Schacter & E. Tulving (Eds.), Memory systems. (pp. 233-268). Cambridge: MIT Press.

Schacter, D.L., Chiu, C.Y.P., & Ochsner, K.N. (1993). Implicit memory: A selective review. Annual Review of Neuroscience, 16, 159-182.

Schacter, D.L., & Graf, P. (1989). Modality specificity of implicit memory for new associations. Journal of Experimental Psychology: Learning, Memory, and Cognition, 15, 3-12.

Slameka, N.J., & Graf, P. (1978). The generation effect: Delineation of a phenomenon. Journal of Experimental Psychology: Human Learning and Memory, 4, 592-604.

Srinivas, K., & Roediger, H.L. (1990). Classifying implicit memory tests: Category association and anagram solution. Journal of Memory and Language, 29, 389-412.

Thapar, A. & Greene, R.L. (1994). Effects of level of processing on implicit and explicit tasks. Journal of Experimental Psychology: Learning, Memory, and Cognition, 20, 671-679.

Tulving, E., & Schacter, D.L. (1990). Priming and human memory systems. Science, 247, 301-306.

Weldon, M.S. (1991). Mechanisims underlying priming on perceptual tests. Journal of Experimental Psychology: Learning, Memory, and Cognition, 17, 526-541.

Weldon, M.S., & Coyote, K.C. (1996). Failure to find the picture superiority effect in implicit conceptual memory tests. Journal of Experimental Psychology: Learning, Memory, and Cognition, 22, 670-686.

Weldon, M.S., & Roediger, H.L. (1987). Altering retrieval demands reverses the picture superiority effect. Memory & Cognition, 15, 269-280.

Author info: Correspondence should be sent to: Dr. Banu Cangoz, Dept. of Psychology, 06532 Beytepe, Hacettepe University, Ankara, Turkey. Email: banucan@hacettepe.edu.tr North American Journal of Psychology, 2005, Vol. 7, No. 3, 389-406. [C] NAJP

Banu Cangoz

Hacettepe University, Turkey
TABLE 1 Proportion of Correct Responses Correctly Solved as a
Function of Solution and Study Type (LOP Effect) Variables

                          Solution Type

Study Type              Single   Multiple

Physical orienting       .63       .51
Semantic orienting       .56       .42
Control                  .22       .31
(nonstudied baseline)

TABLE 2 Proportion of Correct Responses Correctly Solved as a
Function of Solution and Study Type (LOP Effect) Variables

                          Solution Type

Study Type              Single   Multiple

Reading                  .70       .47
Generating               .63       .36
Control                  .35       .20
(nonstudied baseline)
COPYRIGHT 2005 North American Journal of Psychology
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2005 Gale, Cengage Learning. All rights reserved.

 
Article Details
Printer friendly Cite/link Email Feedback
Author:Cangoz, Banu
Publication:North American Journal of Psychology
Geographic Code:1USA
Date:Dec 1, 2005
Words:6461
Previous Article:Autobiographical memory and dysphoria: the effect of mood, gender, and cue type on generality and latency.
Next Article:A preliminary study of the impact of age-related pathologies on cognitive functioning and waking EEG.
Topics:

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters