Abstract
Research shows that divided attention leads to false recognition of encoded information. Such studies rely on the popular Deese-Roediger-McDermott (DRM) paradigm. Participants are usually presented with lists of semantically related words under full or divided attention conditions, after which they are tested for recognition and recall abilities. This study involved 139 (103 females and 36 males) students of a cognitive psychology class based on the DRM approach. Their ages range from 18 to 34, and the mean age is 20.3 years. The participants are divided into two groups and presented with four 10-word lists to listen to, while the second group is instructed to write down random numbers at the video’s beat. After that, the participants are given a recognition test based on word lists. An independent-samples t-test was used to analyze the results. From the results, it was found that divided attention reduces correct and false recognition.
When people study a list of words, they are often susceptible to falsely recalling or recognizing a non-presented critical word (from now on referred to as critical lure (CL)) which has a strong semantic relationship to the words studied. The observation is based on the Deese-Roediger-McDermott (DRM) paradigm pioneered by Deese in 1959 and reviewed by Roediger and McDermott in 1995 (Coane et al., 2020). The DRM remains a commonly used model in memory research, as observed by Coane et al. (2020) and Finley et al. (2017). In investigations involving the DRM methodology, participants study lists semantically or phonetically or associatively related to a single non-presented CLs (Finely et al., 2020; Pardilla-Delgado & Payne, 2017). The CL is falsely remembered at high rates on subsequent recall or recognition tests, often approaching realistic rates. Usually, the robustness of the effect can be intensified by warning participants about this potent memory illusion, although it will reduce but not completely eradicate false memory degrees.
There are other theories whose basis augments DRM studies’ findings, namely, activation-monitoring theory (AMT) and fuzzy-trace theory (FTT). According to AMT, which Roediger and McDermott developed in 1995, processes occurring during encoding and retrieval affect the accessibility of CL and recognition errors (Coane et al., 2020). It implies that participants make errors of source monitoring during recall or recognition tests and falsely regard them as old. The source monitoring is likened to recognition discrimination proposed by Cann and Ross in 1989, which they argue results in increased recall accuracy when the study odor is restored (Isarida et al., 2020). The FTT posits that participants encode two study items: verbatim traces and gist traces (Reyna et al., 2018). Verbatim traces preserve contextual details of particular items, while gist traces refer to the list’s overall subject. The DRM recollection error occurs when participants falsely recall or recognize CLs since they are related to gist traces.
Based on the DRM model and associated theories, this study seeks to determine the effect of divided attention on recognition among two cognitive psychology class. Having reviewed literature related to the subject, this experiment proposed the hypothesis: divided attention will increase the number of items falsely recognized. The groups were instructed to listen to four 10-word lists simultaneously under full and divided attention conditions to achieve the objective.
Method
Participants
The study was conducted on 141 undergraduate students (age range= 18-34, mean age= 20.3 years) during a cognitive psychology lecture they were enrolled in. The mean year of study for the students was the second year. The participants comprised 103 females, 36 males, and one other, while one person declined to provide their demographic information.
Stimuli and Design
Students with names starting with the letter A-L were in Group 1, while those with names started with the letter M-Z formed Group 2. Both groups were to listen to four 10-word lists from a screen in a lecture hall through the Top Hat program or their smartphones and laptops using the video link provided. The words would be read at a rate of one word every two seconds. Group 2 members were instructed to listen to the four 10-word lists, and while doing that, they were to write down a random number between 1 and 20 to the beat they would hear in the video. Both groups were supposed to listen to the recording only once.
A between-subject design was used since different participants were in each group; that is, one group only listened while another also generated random numbers. The independent variable (IV) was the amount of attention available: the divided and full attention conditions. The dependent variables (DVs) were the number of words correctly recalled and the number of words falsely recognized. The participants then did a test in which they were to indicate whether they heard the words in the lists provided. An independent-samples t-test was then used to analyze the participants’ responses.
Procedure
Group 1 was to listen to the lists and not write anything down until they completed listening to the four lists in the video. Group 2 members were to write one random number for each click they heard in the video. They were not supposed to write the same number repeatedly and to follow any pattern. The goal was to create a list of random numbers; after which they would be asked about the lists they listened to. Once all the participants had finished their tasks, they immediately took a test to examine their recall of the words in the four lists.
Results
An independent-samples t-test was used to determine the effect of attention on correct and false recognition. In the correct recognition tests, the full attention group had a mean of 6.68 with a standard deviation of 1.397. In contrast, the random number generation group had a mean of 5.2727 with a standard deviation of 1.819. Since Sig. (2-tailed) is 0.000 (p<0.001), it means that the p-value is greater than 0.001, hence the variance between the two groups cannot be assumed. The value of t = 5.1 is greater than a critical value of 1.96, which means that the difference in the number of correctly recognized words for the two groups is statistically significant.
In the false recognition test, the full attention group had a mean of 4.4 with a standard deviation of 1.786. In contrast, the random number generation had a mean of 4.485 with a standard deviation of 2.047. Since Sig. (2-tailed) of 0.793 is greater than a p-value of 0.001, it is conclusive that the difference invariance across the two groups is not significant hence assumed. The value of t= -0.263 is less than a critical value of 1.96 which means that the differences in the means of falsely recognized words for the two groups are not significant. The sampling process variations could have caused the difference in the means of falsely and correctly recognized words.
Discussion
The high t-value of 5.1 in the correct-recognition test reveals a significant effect of attention on which recognition is greater under full attention than divided attention. The t-value of -0.263 indicated how divided attention greatly affects the rates of recognition. It means that the mean differences of the two groups were not statistically significant. Based on the results, this study has rejected the hypothesis, which states that divided attention will increase the number of items falsely recognized. Other existing studies provide evidence that divided attention negatively affects word recognition.
Dividing attention can also be used to influence recognition processes. Recent studies show that engaging participants in attention-demanding tasks can disrupt retrieval of studied information. Engaging participants in activities such as random number or letter generation during encoding can effectively divide their attention. Nichols and Loftus (2019) state that “…dividing participants’ attention at encoding has been shown to decrease false recognition of critical lures while dividing their attention at retrieval has been shown to increase critical lures” (p. 980). Although Baddeley et al. resolved that “retrieval must be ‘automatic’ and cost-free in terms of attentional resources”, they found out that retrieval latency increased under divide attention conditions (as cited in Cogn, 2018, p. 1262). The automatic retrieval concept was abandoned in subsequent studies that used the continuous reaction time (CRT) tasks as the attention-dividing exercise and consistently found that reaction time slowed significantly during retrieval under divided attention conditions. The findings led to the conclusion that retrieval processes require attentional resources.
The study had limitations such as unbalanced gender representation in that there were more females (103) than males (36). Considering the small participants’ population and a specific age bracket of 18 to 34, the results cannot be generalized for a large population. Since the study was done in a group, there was a possibility of some participants noting the words as they listened. Future studies should focus on the ways to ensure that effective methods of dividing attention are used to maximize participants’ isolation.
References
Coane, J. H., McBride, D. M., & Xu, S. (2020). They feature boost in false memory: The roles of monitoring and critical item identifiability.Memory, 28(4), 481-493. Web.
Finley, J. R., Sungkhasettee, V. W., Roediger III, H. L., & Balota, D. A. (2017). Relative contributions of semantic and phonological associates to over-additive false recall in hybrid DRM lists.Journal of Memory and Language, 93, 154-168. Web.
Isarida, T., Isarida, T. K., Kubota, T., Nakajima, S., Yagi, K., Yamamoto, A., & Higuma, M. (2020). Video context-dependent effects in recognition memory.Journal of Memory and Language, 113, 1-14. Web.
Nichols, R. M., & Loftus, E. F. (2019). Who is susceptible in three false memory tasks? Memory, 27(7), 962-984. Web.
Pardilla-Delgado, E., & Payne, J. D. (2017). The Deese-Roediger-McDermott (DRM) task: A simple cognitive paradigm to investigate false memories in the laboratory.Journal of Visualized Experiments, 119. Web.
Reyna, V. F., Helm, R. K., Weldon, R. B., Shah, P. D., Turpin, A. G., & Govindgari, S. (2018). Brain activation covaries with reported criminal behaviors when making risky choices: A fuzzy-trace theory approach.Journal of Experimental Psychology: General, 147(7), 1094-1109. Web.