Working Memory Capacity: Visual Stimulation Using Digits, Letters, & Words

Working memory is the active memory mechanism within humans that is responsible for storing and manipulating incoming present stimuli (Gray et. al., 2022). The capacity of working memory is especially important to consider as it relates to the successful execution of tasks due to the information processing that takes place within itself (Ochab et. al., 2022). In working memory, visual, auditory, and tactile stimuli are processed; however, human performance is lowered on memory tasks involving multiple modes of sensory stimuli or dual processing tasks. Past research concerning working memory has focused on topics such as: differing information processing tasks and their effects on working memory capacity (Ochab et. al., 2022), the administration and validity of working memory capacity tests (Mielicki et. al., 2018), and individuals’ differences in working memory capacity (Robison & Brewer, 2022), which all help us to answer the question: does working memory capacity differ based on the type and amount of stimuli presented?

In a study investigating working memory capacity and task dependency, researchers examined the temporal organization of working memory data used in four different experimental visual tasks (Ochab et. al., 2022). The experiment consisted of four visual processing tasks including nonverbal (abstract objects requiring global or local information processing) and verbal (words or pseudowords), phases (information encoding and retrieval) and the times of day (day or night) the experiment took place (Ochab et. al., 2022). In the nonverbal semantic task, the stimuli were Polish words matched by semantic similarity (Ochab et. al., 2022). In the phonological task, meaningless pseudowords, characterized by phonological similarity, were used (Ochab et. al., 2022). In contrast, in the global processing task, the stimuli were abstract figures requiring holistic processing, which differed based on the number of overlapping similarities (Ochab et. al., 2022).. Lastly, in the local processing task, the stimuli differed in one specific detail and thus required local processing (Ochab et. al., 2022). After viewing the stimuli in participants’ assigned condition, they were shown a probe. Participants were to choose whether the probe was shown in the previously administered stimuli or not (Ochab et. al., 2022). The probe could be old, new, or a lure. The lures were new stimuli, but were similar to previously administered stimuli (Ochab et. al., 2022). The results of the study showed that the temporal organization of the fMRI differed based on the task as both geometric or fractal characteristics of the stimuli series’ organization and the cross-correlation between brain areas were strongly associated with the brain activity related to one’s engagement in a given task (Ochab et. al., 2022). This study helps to support the argument that working memory capacity could be impacted by executive functions based on the modality of the stimuli. The modality of the stimuli is an important factor to consider when discussing educational implications as well as societal implications due to the impacts modality could have on learning mechanisms (Viesel-Nordmeyer et. al., 2022).

A similar study was conducted investigating visual and tactile stimuli and their relationship to working memory (Bliss & Hämäläinen, 2005). In the visual task of this study, letters were presented on a computer screen and in the tactile task, plastic letters on a board were explored with physical touch (Bliss & Hämäläinen, 2005). Bliss & Hämäläinen found that the amount of incorrect responses increased with increasing memory load in both tasks; however, the total amount of incorrect responses were significantly lower in the visual task than the tactile task (2005). This suggests that the participants remembered purely visual stimuli better than the tactile stimuli. Similarly, in another study conducted researching the effects of computer-assisted instruction and working memory, results showed that students had higher working memory capacity compared to individuals in the inquiry-based learning group (Chevalère et. al., 2021). In the present study, stimuli will only be represented visually and will not be able to be touched.

Another study investigated the individual differences in working memory capacity individuals as well as attention control, and fluid intelligence (Robison & Brewer, 2022). Working memory capacity was measured using 9 different laboratory tasks such as: operation span, reading span, symmetry span, etc. (Robison & Brewer, 2022). Task-unrelated thoughts were measured via thought probes inserted into 2 of the laboratory tasks, and arousal regulation was measured via pupillometry (Robison & Brewer, 2022). The results showed that individuals who had more task-unrelated thoughts exhibited poorer levels of attentional control, lower working memory capacity, and lower fluid intelligence (Robison & Brewer, 2022). These results suggest that there is a relationship between working memory, attention, and fluid intelligence. This is important to note as working memory could play a role in attentional control status as well as fluid intelligence.

Another factor that may influence the capacity of working memory includes mental health, which was investigated in a study done by Yeung & Fernandes with individuals who experience high levels of social anxiety (2019). The experiment revealed that individuals who experienced high levels of social anxiety had lower levels of working memory capacity when asked to recall socially threatening words (Yeung & Fernandes, 2019). The individual differences in working memory may be helpful to investigate due to the applicability of the combination of correlational findings from Robinson & Brewer (2022), the biological findings from Ochab et. al. (2022), and social applications from Yeung & Fernandes (2019) to the way students are taught in the classroom, potential biological predispositions, and overall cognitive functioning.

While the previously mentioned studies have mostly focused on the relationships between working memory and other cognitive and biological processes, a study by Yamaya et. al. investigated how the implementation of attention focused meditation practices can influence working memory capacity (2021). It has been found that focused attention meditation practice can better top-down processing, which has been linked to working memory capacity and important brain structures such as the dorsolateral prefrontal cortex (Yamaya et. al., 2022). In this study, the experimental group of participants took part in one focused attention meditation and brain activity was measured using functional near-infrared spectroscopy (Yamaya et. al., 2022). The participants in the meditation group showed increased blood flow in the dorsolateral prefrontal cortex and increased working memory capacity, which was measured using a reading span test (Yamaya et. al., 2022). The findings in this study also relate to increased attentional control and its positive impact on working memory capacity as found in Robinson & Brewer’s study (2022). The focused attention meditation may serve as a useful tool to increase working memory in the classroom, or in other societal settings.

The mentioned studies have served as precursors to the present study due to their key findings and mentioned implications. These implications are proposed within educational settings as well as overarching modern society (Makmee, 2022), (Yeung & Fernandes, 2019). While many studies have investigated tactile, visual, and semantic stimuli alone, few studies have examined three types of stimuli in one study. The present study investigates the working memory capacity of individuals on three different stimuli types: digits, letters, and words in a memory recognition task. Based on the previous research regarding semantic knowledge and lowered working memory capacity, the word list stimuli are expected to be more challenging for participants to remember (Gray et. al., 2022).

Method

Participants

There were a total of 30 participants (n = 30) in the present study, all of whom were recruited in-person. Of the 30 participants, 18 identified as Female and 5 identified as Male. In terms of race, there were 19 White participants, as well as 4 Mixed participants. The other 7 participants did not respond. Demographic responses for 18 of the participants were collected using a Google Form, while the remaining 12 participants’ demographic data were verbally collected and documented in a spreadsheet. All participants were over the age of 18 and were enrolled at Ripon College. Additionally, 18 of the participants were enrolled in the Cognitive Processes course at Ripon College. Students enrolled in other psychology courses at Ripon College may have received extra credit from their professors for participating in the present study.

Materials & Procedure

Before participating in the study, all participants read and signed an informed consent document. The participants in this study completed the Memory Span lab on the application CogLab (Francis & Neath, 2023). All trials completed for this lab were on a computer. The lab consisted of 30 trials on the computer in which participants were shown a list of digits, letters, or words. The stimuli appeared in a white rectangle with a fixed focal point in the center that flashed in between each stimulus (digit, letter, or word).The lists varied in length from 3 stimuli to 10 stimuli. There were 10 trials for each type of stimuli. To start a new trial, participants clicked the Start Next Trial button located at the bottom of the screen.

After the flashing sequence of digits, letters, or words was completed for each trial, 10 buttons became active underneath the white rectangle. The buttons consisted of digits, letters, or words that were either in the list or not in the previous list. On each trial, participants were asked to recall the items in the same order they were presented. When participants were finished responding, they clicked the Finished Responding button at the bottom of the screen. If participants incorrectly recalled the list, the list length decreased by 1. The highest list length recalled correctly was recorded for each condition. The independent variable in this study was the length of the list consisting of three levels: digits, letters, and words. The dependent variable was the length of the last correctly recalled list. The estimated time to complete this lab was about 35 minutes and all participants were debriefed following their completion of the experiment. The experimental protocol was approved by the Institutional Review Board of Ripon College.

Data Analysis

In order to determine if there was a significant difference between digits, letters, and words in relation to working memory capacity, a Single Factor Repeated Measures Analysis of Variance (ANOVA) was performed using SPSS Statistical Software (IBM Corp, 2020). The three levels of the independent variable include whether the list consisted of digits, letters, or words. The dependent variable was the length of the last correctly recalled list in each condition. The hypothesis for this study was that participants would have the highest list length of recall in the digit condition, the second highest in the letter condition, and the third highest in the word condition.

Results

Researchers found that there was a significant difference between stimuli list type considering digits (SD = 1.223), letters (SD = .994), or words (SD = .980), compared to recalled list length, F(2,58) = 54.848, p < .001, partial eta squared = .654, observed power = 1 (See Figure 1).

Post-Hoc

Due to the significance found within the main effect of the Single Factor Completely Repeated Measures ANOVA, post-hoc Bonferroni pairwise comparison tests were performed. The follow-up test found that there was a significant difference in recalled list length between digit stimuli and letter stimuli, as participants remembered significantly more digits than letters, p = .021. In addition, there was a significant difference in recalled list length between digit stimuli and word stimuli, as participants remembered more digits than words, p < .001. Lastly, there was a significant difference in recalled list length between letter stimuli and word stimuli, as participants recalled more letters than words, p < .001.

Discussion

The results of this study support the hypothesis that individuals would have the highest amount of working memory capacity for digits, second highest for letters, and third highest for words. These findings support the previous research conclusions such as the study Ochab et. al. (2022) conducted investigating multiple modes of stimuli. In their study, there were significant differences between the varying stimuli groups in terms of working memory capacity and regional brain activity. In our study, there were three different stimuli, which could have resulted in increased activity in different regions of the brain, or cognitive processing in different brain regions. These physical changes measured in the previous study may have an effect on the working memory capacity when viewing digits, letters, or words. In the Bliss & Hämäläinen (2005) study, they found that visual stimuli resulted in fewer incorrect responses when performing a working memory recall task than in tactile stimuli conditions, which may also be associated with cognitive processing in different regions of the brain due to the stimuli’s modality. Based on the previous research by Gray et. al. (2022) regarding semantic knowledge and lowered working memory capacity, the word list stimuli was expected to be more challenging for participants to remember. The present results also support this research as participants had the worst working memory capacity when presented with words. This may be due to the semantic knowledge associated with the words presented, or the amount of letters in each word.

Other factors that may have influenced our results include the findings in Yamaya et. al. (2022), which found that attentional control plays an important role in working memory capacity; high level attentional control resulted in increased levels of working memory capacity. When completing our study, participants were not given any tools to increase their attentional control compared to the focused attention meditation facilitated in the Yamaya et. al. (2022) study. The results of the present study may have resulted in overall higher working memory capacity data in each condition if participants were asked to complete a similar meditation task before completing the experimental task. This idea offers a future opportunity for further research.

The Gray et. al. (2022) study supports the hypothesis that working memory as a whole is a significant predictor of not only what has already been learned, but also what is actively being learned. This is important because learning could possibly be improved if working memory is optimized, which leads to the educational implications of the present research. The participants in this study had better working memory capacity for digits and letters in comparison to words. It is important to consider that the working memory capacity for letters should be optimized before learning words as one cannot interpret a word without remembering what letters make up the respective word. This implication is especially important for educators working with children who are beginning to learn how to read and make meaning of texts. In addition, the working memory capacity for digits may be a predictor of later mathematical academic achievement as numbers will start to increase to more than one digit. Students may first need to have the working memory capacity to remember a single digit before interacting with larger numbers.

Another implication of this research within education are the individual differences seen between subjects. While the majority of participants exhibited the highest level of working memory capacity in the digit condition, there were participants who showed varying degrees of memory capacity in this condition as well as the others. This connects back to the study done with individuals who have high levels of social anxiety, as they showed overall lower levels of working memory capacity compared to individuals who had low levels of social anxiety (Yeung & Fernandes, 2019). For students with high levels of social anxiety, they may struggle to learn in an optimal fashion due to the restraints of their working memory capacity. In addition, their overall scores in the present study may have been affected by their level of social anxiety.

Other limitations that may have played a role in the results of the present study include the sample size, gender proportion, age and race. Due to the small sample size (n = 30) in this study, external validity and reliability of the experiment may be lowered. In addition, the proportion of different genders was not equal, which may have skewed the results due to a higher number of female participants. All of the participants were between the ages of 18 and 21, which limits the applicability of these findings to other age groups. Lastly, there were unequal proportions of race as the majority of participants were Caucasian, which limits the applicability of the present study to other racial groups.

In conclusion, the results of this study support the original hypothesis that individuals would have the highest amount of working memory capacity for digits, second highest for letters, and third highest for words. The present study used the length of the last correctly recalled list of stimuli to measure working memory capacity and the lists were manipulated by length and type of stimuli. Results suggest that working memory capacity varies based on the type of stimuli presented as well as the amount of stimuli presented within a given trial. The results of this study can be applied to changes in the education field as well as jobs that require optimized working memory capacity. Further research is suggested measuring attentional control levels and different types of stimuli as well as more research on the individual differences present within working memory capacity.

References

Bliss, I., & Hämäläinen, H. (2005). Different working memory capacity in normal young adults for visual and tactile letter recognition task. Scandinavian Journal of Psychology, 46(3), 247–251. https://doi.org/10.1111/j.1467-9450.2005.00454.x

Chevalère, J., Cazenave, L., Berthon, M., Martinez, R., Mazenod, V., Borion, M.-C., Pailler, D., Rocher, N., Cadet, R., Lenne, C., Maïonchi-Pino, N., & Huguet, P. (2021). Computer-assisted instruction versus inquiry-based learning: The importance of working memory capacity. PLoS ONE, 16(11), 1–23. https://doi.org/10.1371/journal.pone.0259664

Francis, G., & Neath, I. (2023). Enhanced CogLab [Online program] (Version 5). Cengage Learning. https://coglab.cengage.com

Gray, S. I., Levy, R., Alt, M., Hogan, T. P., & Cowan, N. (2022). Working Memory Predicts New Word Learning Over and Above Existing Vocabulary and Nonverbal IQ. Journal of Speech, Language & Hearing Research, 65(3), 1044–1069. https://doi.org/10.1044/2021_JSLHR-21-00397

IBM Corp. (2020). IBM SPSS Statistics for Windows (Version 27.0) [Computer software]. IBM Corp.

Mielicki, M. K., Koppel, R. H., Valencia, G., & Wiley, J. (2018). Measuring working memory capacity with the letter–number sequencing task: Advantages of visual administration. Applied Cognitive Psychology, 32(6), 805–814. https://doi.org/10.1002/acp.3468

Ochab, J. K., Wątorek, M., Ceglarek, A., Fafrowicz, M., Lewandowska, K., Marek, T., Sikora-Wachowicz, B., & Oświęcimka, P. (2022). Task-dependent fractal patterns of information processing in working memory. Scientific Reports, 12(1), 1–14. https://doi.org/10.1038/s41598-022-21375-1

Makmee, P. (2022). Increasing Attention and Working Memory in Elementary Students Using Mindfulness Training Programs. FWU Journal of Social Sciences, 16(3), 107–119. https://doi.org/10.51709/19951272/Fall2022/8

Robison, M. K., & Brewer, G. A. (2022). Individual differences in working memory capacity, attention control, fluid intelligence, and pupillary measures of arousal. Journal of Experimental Psychology. Learning, Memory & Cognition, 48(9), 1296–1310. https://doi.org/10.1037/xlm0001125

Viesel-Nordmeyer, N., Röhm, A., Starke, A., & Ritterfeld, U. (2022). How language skills and working memory capacities explain mathematical learning from preschool to primary school age: Insights from a longitudinal study. PLoS ONE, 17(6), 1–25. https://doi.org/10.1371/journal.pone.0270427

Yamaya, N., Tsuchiya, K., Takizawa, I., Shimoda, K., Kitazawa, K., & Tozato, F. (2021). Effect of one session focused attention meditation on the working memory capacity of meditation novices: A functional near infrared spectroscopy study. Brain & Behavior, 11(8), 1–11. https://doi.org/10.1002/brb3.2288

Yeung, R. C., & Fernandes, M. A. (2019). Altered working memory capacity for social threat words in high versus low social anxiety. Anxiety, Stress & Coping, 32(5), 505–521. https://doi.org/10.1080/10615806.2019.1626838

Figure 1

Mean Lengths of Correctly Recalled Lists in Each Condition

Note. This table displays the mean length of the last correctly recalled list in the digit, letter, and word conditions. The bars represent the standard error. The digit condition had a significantly higher amount of stimuli recalled than both the letter and word conditions. The letter condition also had a significantly higher amount of stimuli recalled than the word condition.