Assessing the Relationships Shared by Different Tasks Measuring Working Memory Capacity - Lab Report Example

? Working Memory Experiment: Assessing the Relationships Shared by Different Tasks Measuring Working Memory Capa This experiment was carried out to test the relationships between different tasks that measure Working Memory capacity. It was proposed that all four tasks – MU, OS, SS and SSTM – would be significantly correlated and that MU, OS and SS would share stronger relationships among themselves than with SSTM. Data was collected from 444 students, of which data on 406 was deemed acceptable for analysis. The correlations computed suggested that both the hypotheses could be accepted, and that although all WM capacity tasks did share similarities, they did also indicate the existence of differences in processing of verbal and spatial information. Introduction Human memory has been of interest to both, theoretical and experimental psychologists for a long time. A traditional model of memory suggested that individuals have a passive short term store (STM) where information is rehearsed, and then sent to a long term store. Over time, the description of the STM was found to be lacking, and Baddeley and Hitch (1974) proposed the concept of the Working Memory (WM). This refers to an individual’s ability to not only rehearse but also to manipulate information during a short period of time. They suggested that the WM consisted of a processing system called the central executive that oversees the working of two slave systems – the phonological loop and the visuospatial sketchpad. The phonological loop processes verbal and numerical information, while the visuospatial sketchpad processes spatial information. The central executive is believed to incorporate the information from both these sources to aid the overall information processing (Baddeley & Hitch, 1974). This suggests that the WM is a far more active component as compared to the concept of the STM. It has also been found to be a more valid predictor of individual performance on different tasks as compared to the STM (Daneman & Carpenter, 1980). Since the definition of WM includes both passive storage and active processing, it is better able to explain complex cognitive processes. According to Daneman and Carpenter (1980), WM capacity can help explain differences in reading. Solving of mathematical problems as well as the individual differences in doing so can be explained using WM (Hitch, Towse & Hutton, 2001). It is believed that differences in WM capacity may explain differences in reasoning (Kyllonen & Christal, 1990) and attentional control (Kane, Bleckley, Conway, & Engle, 2001). Thus, measuring the capacity of WM can help explain differences in intelligence and ability as well (Kyllonen & Christal, 1990). A number of different tasks have been used to measure WM capacity. Lewandowsky, Oberauer, Yang and Ecker (2010) have demonstrated the utility of four tasks that measure different processes in WM including spatial reasoning, numerical ability, verbal ability, and updating of information. Schmiedek et al. (2009) suggest that updating of information uses the same processes as storing information in WM. Conway et al. (2005) believe all methods of measuring WM capacity share similar features, while other results by Oberauer (2005) suggest that the relationships shared by verbal and spatial tasks is not perfect. Based on these finding, the following hypotheses were proposed: All four tasks will be significantly related to one another to indicate that they all measure WM capacity. Tasks associated with the use of the phonological loop (MU, OS and SS) will share stronger relationships with each other than they will with SSTM. Method An experimental study was conducted to verify the relationships shared by four WM tasks similar to those described by Lewandowsky et al. (2010). The study used a correlational design to test the different hypotheses. Participants Initially, data was collected for all four tasks from 444 participants. All participants were college students taking the 2215 and the 3315 courses. The age and gender of participants were not considered as the same participants provided data for all tasks. Based on distractor task performance, 34 participants were excluded for performance less than the chosen requirement of 75% accuracy. This was done to ensure that participants did not ignore the secondary tasks. Of the remaining 410 participants, 4 were outliers whose data was also excluded to remove any confounding effect. Thus data from 406 participants was used in the final analysis. Tasks Memory Updating (MU): In this task, participants were presented with three to five frames concurrently per trial with a number in each frame that was updated with mathematical operations. The numbers were presented for one second, and the updating steps for 1.3 seconds each. The operations ranged from +7 to -7 to yield a final result between one and nine. The individual had to recall the number at the end of a series of such steps. The task had 15 trials, with each trial having between two to six updating steps that were presented identically for all participants. All correct responses per trail were considered, and the final scores for each trial were then aggregated. Operation Span (OS): This task required the participant judge presented mathematical operations whose operands ranged from one to ten as being correct or incorrect while simultaneously holding a set of alphabets in mind. Each operation was followed with a to-be-remembered (TBR) alphabet. Half of the mathematical operations were correct and half were incorrect and all answers were greater than zero. These were randomly presented for up to 3 seconds (or till the participant responded) while TBR alphabets were presented for one second. Participants received 15 trials with list length ranging from four to eight; with three trials per list length. Sentence Span (SS): This task was similar to the OS task, but used 90 sentences that were to be judged as either meaningful or not meaningful instead of mathematical operations. Half the sentences were meaningful while the other half were not. Each sentence was presented up to four seconds, and the TBR letters for 1 second each. The participant received 15 trials of list lengths varying between 4 – 8 sentences with three trials per list length. Spatial Short Term Memory (SSTM): For this task the participants had to remember the pattern formed by two to six sets of dots within a grid. The sequence of presentation and the actual position of the dots were not considered; only their spatial relation to each other was. The participants were all exposed to 30 trials in the same order so that there were six trials per set size. Each dot was presented for 0.9 seconds with one second intervals between dots. Scores across all trials were aggregated. Procedure The four tasks were administered using computers, and all participants were exposed to the different tasks in the same order. The instructions for each task were presented before the task, doubts of participants were clarified and then the trails were conducted. Aggregate scores across all trials were calculated per participant for each task. Results Data collected from 406 participants was used to test the hypotheses. The mean scores and the standard deviations for all participants on the different tasks were tabulated in table 1. These scores suggest that participants performed reasonably well on all tasks. Performance seems to be similar on MU, OS and SS, while it is higher on SSTM as compared to the others. The table also contains the group mean and SD for two secondary judgment tasks. Table 1: Means and Standard Deviations on the tasks Variables Mean SD MU 0.659 0.167 OS 0.722 0.110 SS 0.704 0.130 SSTM 0.849 0.052 OS (pt) 0.923 0.052 SS (pt) 0.921 0.044 In order to test the hypotheses, correlations were computed between the scores of the individual on the four tasks. All four tasks were significantly correlated with each other. The two span processing tasks (OS and SS) shared the strongest relationship, which qualified as a high correlation. This was followed by the relationships shared by the relationship between OS and MU and the relationship between SS and MU both of which were moderate relationships. The relationship between SSTM and MU was a low to moderate one, while the relationship between SSTM and OS and the relationship between SSTM and SS were both significant but low relationships. Table 2 summarizes the relationships between the variables. Table 2: Correlations (DF = 405) MU OS SS SSTM MU 1.000 0.534* 0.525* 0.354* OS 1.000 0.721* 0.239* SS 1.000 0.206* SSTM 1.000 *p < 0.05 These results show that the proposed hypotheses may be accepted. Discussion This experiment was conducted to test the relationship between four different tasks that are used to measure the capacity of WM. It also tried to verify if the tasks associated with verbal processing that are expected to use the phonological loop and those that require spatial processing (and are expected to use the visuospatial sketchpad) were similarly processed by participants. The results suggest that both the hypotheses proposed by the study may be accepted. All four tasks were significantly correlated to each other. This suggests that these tasks share a single underlying factor (Conway et al., 2005). Thus, the capacity for processing of information in WM is similar in any given person across different tasks. This is also demonstrated by the mean scores for the different tasks that suggest that people have similar processing capacity for MU, OS and SS tasks. It may be that the higher processing capacity demonstrated for the SSTM task is due to the fact that verbal tasks that measure WM capacity are known to show proactive interference (Lustig, May, & Hasher, 2001). Thus, as the participant does through the trials, they may experience interference of verbal tasks that reduces accuracy for later trails. This may not happen as easily for spatial tasks. But while these correlations are significant, they are not very high (with the exception of the relationships between OS and SS). This suggests that each of the tasks does indeed measure a different type of processing. Particularly, SSTM – which measures spatial processing – has low correlations with the other three tasks (all of which measure different kinds of verbal processing. These results suggest that although both verbal and spatial processing share a common denominator (Schmiedek et al., 2009; Conway et al., 2005) they also have important differences (Oberauer, 2005). Baddeley and Hitch (1974) suggested that the phonological loop and the visuospatial sketchpad both use different means to process information. MU, OS and SS tasks are all associated with the use of the phonological loop (verbal processing) while the SSTM task is expected to use the visuospatial sketchpad. Thus, these results support toe assumption that the two slave systems do process information differently, so that people may show differing capacity for processing verbal and spatial information. These finding may also explain the differences seen in an individual’s spatial and verbal abilities as seen in intelligence and ability testing. References Baddeley, A. D., & Hitch, G. (1974). Working memory. In G. A. Bower (Ed.), The psychology of learning and motivation (Vol. 8, pp. 47-89). New York: Academic Press. Conway, A. R. A., Kane, M. J., Bunting, M. F., Hambrick, D. Z., Wilhelm, O., & Engle, R. W. (2005). Working memory span tasks: A methodological review and user's guide. Psychonomic Bulletin & Review, 12 (5), 769-786 Daneman, M., & Carpenter, P. A. (1980). Individual differences in working memory and reading. Journal of Verbal Learning & Verbal Behavior, 19, 450-466. Hitch, G. J., Towse, J. N., & Hutton, U. (2001). What limits children’s working memory span? Theoretical accounts and applications for scholastic development. Journal of Experimental Psychology: General, 130, 184-198. Kane, M. J., Bleckley, M. K., Conway, A. R. A., & Engle, R. W. (2001). A controlled-attention view of working-memory capacity. Journal of Experimental Psychology: General, 130, 169-183. Kyllonen, P. C., & Christal, R. E. (1990). Reasoning ability is (little more than) working-memory capacity?! Intelligence, 14, 389-433. Lewandowsky, S., Oberauer, K., Yang, L.-X., & Ecker, U. K. H.(2010). A working memory test battery for MatLab. Behavior Research Methods, 42, 571–585. Lustig, C., May, C. P., & Hasher, L. (2001). Working memory span and the role of proactive interference. Journal of Experimental Psychology: General, 130, 199-207. Oberauer, K. (2005b). The measurement of working memory capacity. In O. Wilhelm & R. W. Engle (Eds.), Handbook of understanding and measuring intelligence (pp. 393-408). Thousand Oaks, CA: Sage. Schmiedek, F., Hildebrandt, A., Lovden, M., Lindenberger, U., & Wilhelm, O. (2009). Complex span versus updating tasks of working memory: The gap is not that deep. Journal of Experimental Psychology: Learning, Memory, & Cognition, 35, 1089-1096. Read More