Introduction
This empirical study was done by Jannati, McDonald, and Di-Lollo at Simon Fraser University. The article explores distinct disparities in the pace of processing as compared to K scores of VSTM capacity. The article begins by exploring estimation techniques for visual short-term memory (VSTM) capacity in K scores. The article is well researched because it contains an empirical study of visual short-term memory capacity. The article begins by giving a brief summary of its experiments and findings in the abstract section. The authors show through empirical study that K is not only affected by size but also by the speed of processing. The authors argue that a study on the rate of encoding shows a significant increase in K scores for fast encoders. Therefore, the article shows that the rate of encoding is also a significant factor in the estimation of K scores. This paper will provide a brief summary of the article (Jannati, McDonald, & Di-Lollo, 2015).
According to the authors, VSTM grasps conceptual illustrations of few items actively for some time. The article also argues that estimates of the capacity of VSTM are done in terms of K scores. The authors introduce a change detection paradigm, which is utilized to obtain K scores. In a change detection paradigm, the article outlines that a concise memory array with certain items is usually trailed by an assessment array after a retention interval. In this case, the observer’s job is to note the difference or similarity of the two arrays (Jannati, McDonald, & Di-Lollo, 2015). The authors approximate that 3-4 items have been registered as K scores using a change detection paradigm in the past. In this section, the authors provide an in-depth literature review on estimates of the capacity of VSTM. Moreover, the authors synthesis their idea on the influence of the rate of encoding on VSTM capacity. They argue that for fast observers, processing speed is also significant in the estimation of VSTM capacity (Habekost & Starrfelt, 2009).
In this section, the authors argue that K values are not only determined by VSTM capacity but also by processing speed. In supporting their arguments, the authors conflict past research, which estimated that most processing of memory array could be accomplished in 100ms. In fact, they argue that processing could go on for hundreds of milliseconds if there were no trailing masks. The authors go on to prove their point by citing a study by Vogel and Machizawa, which brought to light the correlation between VSTM and the number of items held. In the process, they show that multiples of 100 milliseconds elapsed for the greatest amplitude to be attained. In essence, memory could still be processed after display offset (Jannati, McDonald, & Di-Lollo, 2015).
Experiments and results
The authors performed two experiments to prove empirically that individual differences in the rate of encoding also predicted K scores. Experiment 1 concerning the inclusion of a temporary trailing masking stimulus. This experiment was done to test Luck and Vogel’s conclusion. It was clear that a trailing mast wasn’t used by Vogel andLuck. According to the authors, this must-have interfered with their conclusions. The second experiment explored the possibility of slower encoders displaying lower K scores than faster encoders (Jannati, McDonald, & Di-Lollo, 2015). Essentially, the second experiment was done to prove that the speed of processing was significant in the approximating capacity of VSTM (Woodman & Vogel, 2008).
Experiment One
This experiment involved 48 participants from the University. Every participant sat in a dimly lit room 60cm from the monitor that refreshed at 120hz.The observers performed change detection tasks as directed. Stimulus presentation was regulated using software known as E-Prime. Moreover, response gathering was also checked using the software (Jannati, McDonald, & Di-Lollo, 2015). Stimuli contained a test array, a mask and a memory array. In each trial, memory array were arbitrarily chosen such that no color would appear more than once in the display. The observers point out if the array was similar or dissimilar using either of the push buttons of mouse. When utilized, the covers contained 4 painted quadrangles forming two by two models. Moreover, the colors in the masks were randomly chosen. In essence, a mask was utilized at every square in the memory array. Every participant performed 480 tests; that is, 90 every condition for the 5 conditions and additional 30 for 100-no mask condition (Curby & Gauthier, 2009).
Results and discussion for experiment 1
According to the authors, they performed pair wise comparison between appropriate conditions. Moreover, they performed ANOVA on the first four conditions, which showed that major connections and factors were meaningful as observed in p values ≤ 0.005. However, they also noted that some of their observations were open for scrutiny especially those concerning functions that were observed to be near ceiling. Additionally, they executed numerical tests on five discrete projections as given above. Condition 100-no mask proved similar in correct response to condition 500-no mask.
Additionally, it was found that performance in 100-no mask was considerably superior to that of 100-mask (Jannati, McDonald, & Di-Lollo, 2015). This response rejected past hypothesis, which argued that processing could be accomplished in around 100ms. Additionally, it was found that condition 100-ISI-mask was higher than that of condition 100-mask. Furthermore, performance in condition 100-ISI-mask was found to be analogous to that of 500-mask. Performance in condition 500-no mask was found to be significantly higher than that of condition 500-mask. In general, these results showed that estimation of VSTM capacity using change detection paradigm encountered temporal limitations that unmasked rate of encoding as a factor (Jannati, McDonald, & Di-Lollo, 2015).
Experiment 2
As mentioned earlier, the second experiment was done to show that the rate of encoding influenced estimation of VSTM capacity. The authors observed encoding speed in one task. Additionally, they approximated K using another distinct task. Firstly, K was estimated using change detection pattern while encoding speed was approximated using backward-masking method. 110 participants acted as observers during the study. The second experiment was similar to the first although observers also performed backward-masking tasks. This task also involved PEST runs from which ISI was obtained. It should be noted excessive approximations were avoided by using median values (Jannati, McDonald, & Di-Lollo, 2015).
Experiment 2 results and observations
K score average was 2.6 with a standard deviation of 0.82. The ISI score was estimated as 74ms (SD 49). In this regard, about 91ms were necessary for encoding 2 letters in the system. This result agreed with Vogel’s estimation, which estimated the duration of encoding one letter as 50ms. The authors divided participants into two main groups, which represented swift encoders and slow encoders. This was achieved by using ISI scores of the given participants. These results were compared to the K scores from the two groups. The researchers found that estimated K scores for observers in fast group were considerably higher than that of slow group (2.83 vs. 2.37). Furthermore, another look at the statistical correlation between the two scores was considerable. This indicated that faster rates of encoding signified higher K scores. According to the authors, this showed that K scores indicated both rate of encoding and capacity of VSTM (Jannati, McDonald, & Di-Lollo, 2015).
Discussion
The authors stated that this article was intended to inspect the infamous view that K determined VSTM capacity only. Moreover, the study was designed to scrutinize the impression that approximation of VSTM capacity was impervious to restrictions in processes. According to the authors, the first trial demonstrated that sequential limits affected approximation of capacity of VSTM using the first paradigm. This revelation exposed the role of speed of processing. Additionally, the authors argued that experiment 2 shadowed the verdict in experiment 1 by showing that slower encoders registered lower K scores than faster encoders. The results exhibited the fact that K scores not only reflect capacity of VSTM but also the rate of encoding into VSTM (Bundesen, Habekost & Kyllingsbæk, 2011).
Conclusion
According to the authors, their study focused on three central points. Firstly, it was aimed at disconfirming Luck and Vogel’s conclusion that K scores estimation using change detection paradigm was exclusively based on VSTM capacity. Instead, the study confirmed that change detection paradigm was constrained by temporal limitations. Additionally, it was intended to demonstrate that size of VSTM found using the first paradigm was associated with the processing speed. Additionally, the study increased the possibility of encoding speed being a factor in investigating the connection between other constructs and VSTM system.
Reference List
Bundesen, C., Habekost, T., & Kyllingsbæk, S. (2011). A neural theory of visual attention and short-term memory (NTVA). Neuropsychologia, 49, 1446–1457.
Curby, K. M., & Gauthier, I. (2009). The temporal advantage for individuating objects of expertise: Perceptual expertise is an early riser. Journal of Vision, 9, 7–13.
Habekost, T., & Starrfelt, R. (2009). Visual attention capacity: A review of TVA-based patient studies. Scandinavian Journal of Psychology, 50, 23–32.
Jannati, A., McDonald, J., & Di-Lollo, V. (2015). Individual Differences in rate of Encoding Predict estimates of Visual Short-Term Memory Capacity (K). Canadian Journal of Experimental Psychology, 69(2), 212-220. Web.
Woodman, G. F., & Vogel, E. K. (2008). Selective storage and maintenance of an object’s features in visual working memory. Psychonomic Bulletin & Review, 15, 223–229.