{br} STUCK with your assignment? {br} When is it due? {br} Get FREE assistance. Page Title: {title}{br} Page URL: {url}
UK: +44 748 007-0908, USA: +1 917 810-5386 [email protected]


Investigating Correlation between Sentence Prediction and Recognition Memory among University Students


Subject Research Report Pages 6 Style APA


Research Report


The research used a sample population of 344 students though only 313 participants met the selection criteria. They consented to engaging in the research which was approved by the university’s ethics committee. The actual research was done in two phases; the first stage also known as encoding phase involved sentence prediction task while the second phase, also referred to as production/ recognition memory task tested the ability of the participants to recall. The research participants were presented with a list consisting of 192 words during the recognition memory task/ phase and were asked to encode the words. Other criteria were also set to ensure the tests were standardized and that only the required data was collected. The performance exhibited in recognizing the new and old words presented a definitive measure of the recognition memory performance of the research participants. The data was used to compute the d prime for each encoding condition where the ability for the participants to discriminate between new and old words in standard deviation units was established. The findings showed that there was no significant difference between the two conditions generated. These findings lead to questioning on the factors that influence the recognition memory among university students.

Investigating Correlation between Sentence Prediction and Recognition Memory among Students

Determinants of Sentence Prediction and Recognition Memory

Recognition is the most widely studied section of declarative memory. According to Stern and Hasselmo (2009), memory denotes the ability for an individual to correctly judge a recent encounter as having been experienced or presented previously. In such cases, recognition memory is composed of two components; familiarity and recollection. Given the complexity of understanding recognition memory, the topic has attracted extensive research studies as psychologies try to undertint the human brain. Recognition memory has been tested using different paradigms to ascertain the ability to memorize and recall.

One of the classical approaches to testing recognition memory is by presenting the subjects with a stimuli to be learnt after which the participants are exposed to a variable delay period. The period is aimed at enabling them internalize the stimuli after which the experimenter presents the subjects with the previous stimuli mixed with new stimulus also known as lures. The subjects are then asked to respond with a no or yet to each of the stimulus. This stimulus test is known as a yes/no recognition test and has been proven effective for testing recognition memory. The subjects have to correctly identify while stimulus were presented previously and which ones are the lures (Stern & Hasselmo, 2009). Such tests are often known as the forced choice tests for recognition memory. These tests cannot be used effectively on people suffering memory disorders as they will respond with a no to each of the stimuli. To avoid response bias, it is important to properly select the research population for the research. In addition, using the forced choice tests have proven less sensitive to biased responses thus, they can be used to sensitively assess impairments in recognition memory and the resultant response biases.

Recent studies have evidenced that the yes and no responses provide more accurate predictions of ability for a person to recollect, familiarize or recognize items seen before. In such tests, the ‘yes’ represents the ability for the person to familiarize and recollect while a ‘no’ refers to inability to recall or familiarize. In some cases, the experimenter might be required to distinguish between familiarity and re-collective responses. The difference between a weaker sense of familiarity and strong feeling of recollection are often known as the ‘know’ versus ‘remember’ distinction and are important experimental tests that predict recognition memory. Conducting these tests justifies findings by Kane, Brown, McVay, Silvia, Myin-Germeys and Kwapil, (2007) noting that human beings have varied cognitive abilities. This statement is backed by conventional intelligence tests which justified that people have different cognitive abilities which influences their life experiences either positively or negatively.

According to Kane et al. (2007) people with high intelligence have the tendency of score better grades, securing prestigious occupations, attaining more knowledge and education, have low risk of incarceration, and also, they are less likely to killed by automobile accidents. The author further notes that people with high intelligence show an openness to new experiences since they are novelty seekers, curious, unconventional and aesthetically sensitive than those with lower intelligence. These sentiments are backed by cognitive-mechanistic theories of intelligence which argue that differences in processing speed, sensory discrimination, recognition memory, and working memories are all effective predictors of the effectiveness of a mental system of a person. The more effective the mental system, the lesser the dramatic consequences experienced by the person. Because of these findings, it is certain that recognition memory is a key predictor of a person’s quality of life. It is therefore important to evaluate how recognition memory impacts sentence prediction. Guided by this rationale, this paper is guided by the hypothesis that recognition memory is a key determinant of an individual’s ability for sentence prediction.

Figure depicting d prime data

Figure 1: d prime data predicting encoding condition of the sentence prediction task


The results showed that when all the research participants are exposed to the list consisting of 192 words during the recognition memory phase, their classification of the worlds is different depending on their encoding condition during the sentence prediction task. The three classifications namely; generated and produced, generated and read and not generated but read, were used to predict the encoding condition of the participants. The findings showed that most of the words were likely to be generated and produced. The range of production was rated at 9 to 31 meaning that out of 31 people, only 9 had the ability to generate and produce old items. This figure describes the memory performance of the research participants. As shown in figure 1 above, the d’ shows the discrimination between the new and old words. Generally, 0.057 subjects could effectively discriminate between the new and old words that could be generated and produced.

A lower proportion of the study population amounting to 0.055 could further discriminate the new and old words that could be generated and read while a similar proportion of respondents could discriminate between the old and new words that were not generated but read. It is surprising that an analysis of the full sample of research participants (n = 313) resulted into an exact data pattern as that of the restricted sample. In this case, the restricted sample contained a lesser population of 205 participants. This was after 14 participants were eliminated from the analysis since they failed to understand the research instructions while others were removed for failing to acknowledge that their research data can be analyzed. Notably, the restricted sample was still a meaningful sample population which enabled the research to accurately detect significant effects of encoding condition on sentence prediction and recognition memory. Therefore a restatement of these results in terms of the hypothesis predicts that there is no significant correlation between sentence prediction and recognition memory.

There is no significant correlation between recognition memory and sentence prediction

A research article by Hubbard, Rommers, Jacobs and Federmeier (2019) report that human beings process language, they often use contexts to make predictions of upcoming information. Besides, context is used to augment understanding and to influence processing as seen in neural and behavioral tests. Despite the fact that countless research studies have shown an immediate facilitative impacts of confirmed predictions, downstream consequences resulting from prediction are still less explored. As a result, Hubbard et al. (2019) examined the downstream consequences of recognition memory for words after the subjects were exposed to a number of words and sentences. In the research, the subjects read weakly and strongly constraining sentences attached to either unexpected or expected endings.  The participants were then tested for the ending of the sentences and words. One of the researcher added his name as part of the list. Later on, they tested the ability of the subjects to remember the sentence endings. The list contained words that could be predicted yet they were never read.

Behaviorally, the research participants showed that they could successfully discriminate between the new and old items but they had higher chances of raising false alarm to items introduced to lure them. This observation validates that predicted words can still remain in the human memory even after the predictions were disconfirmed. Additionally, it was observed that even when the rates of false alarms differed in comparison to the constraints, all event-related potentials (ERPs) varied significantly between the words that were weakly predictable and those that created strong false alarms (Hubbard et al. 2019). It was further noted that words with unexpected endings produced bigger LPC magnitudes. This finding suggests that which suggests that the research participants allocated more attention to these specific words and had a higher occurrence of episodic recollection. In contrast, extremely foreseeable endings of sentences produced lower LPC amplitudes during the memory tests. Citing this correlation, the research revealed that predicting could lead to false memory consequently reducing the memorization or recollection of predictable information. These findings justify that there is no correlation between recognition memory and sentence prediction.

A study of executive functions shows that recognition memory is dependent on multiple factors thus not an accurate measure of sentence prediction

According to Miyake and Friedman (2012), executive functions (EF) refer to a set of control processes used for general purposes to regulate the thoughts and behavior of a person. EFs has become a common research topic in the recent past. The concept is studied across the various sub-disciplines of psychology. After a detailed research, Miyake and Friedman (2012) summarized their findings on EF which helps highlight the correlation between sentence prediction and performance of recognition memory. In their research, the researchers sought to understand how individual differences in executive functions were correlated to biological and cognitive underpinnings. The authors then developed a unity/diversity framework that explains four conclusions that were made from their research study.

The study showed that individual differences in EF, as measured using laboratory tasks evidenced that diversity and unity are separable yet still correlated. The findings further showed that recognition memory was determined by genetic contributions. The third finding was that individual differences in EF were related to diverse societally and clinically important phenomena. Fourth, the executive functioning in some individuals was a sign of developmental stability (Miyake & Friedman, 2012). A critical analysis of these findings has a significant influence on the expected results for this research paper. The findings rationalize that there are many factors that contribute towards a person’s ability to correctly make old and new predictions of sentences. Perhaps these sentiments further justify the need to study how the two variables of sentence prediction and recognition memory are correlated.

Unconscious influences play a role in sentence prediction and not recognition memory

Jacoby and Kelley (1992) introduce the Freudian slip approach in explaining the impact of unconscious influence on sentence prediction. The concept of Freudian slip argues that unconscious influenced result from errors and thus, it should be placed in opposition to intentional or consciously controlled memory. In such cases, the authors propose the use of a fame judgement experiment they had done some years back. The test results showed that subjects who had undertaken the study with full attention could easily recognize the names that appeared and reappeared on the list of the famous people and thus, easily identify whether the names were famous or non-famous. The findings showed that there was a higher probability of classifying non-famous people as famous and vice versa when the subjects showed divided attention to details (Jacoby & Kelley, 1992). This phenomenon was described as the divided attention condition and was blamed on unconscious influence of the recognition memory.  These findings justify that unconscious influences impact the recognition memory and its ability to make correct sentence predictions.

Limitations of study design

The main internal validity challenge of the study design was the short exposure of the subjects to the sentences to be predicted. As a result, future research design could consider undertaking a longitudinal study to evaluate whether the results will be varied if the respondents were tested over a prolonged period of time, say three years.


The research seeks to analyze the correlation between sentence prediction and recognition memory. The discussions show that there is no significant correlation between the two variables. After observing event-related potentials (ERPs), the researchers identified that a person’s ability to memorize and recollect predictable information was a major determinant in sentence prediction than recognition memory. Additional research focusing on executive function showed that recognition memory was dependent on multiple factors, therefore, it could not be an accurate determinant of the ability to make correct sentence predictions. Another finding that influences the outlook of the research study is that unconscious influences play a significant role in sentence prediction. This particular findings leads to questioning on whether sentence prediction is dependent on recognition memory.





 Hubbard, R. J., Rommers, J., Jacobs, C. L., & Federmeier, K. D. (2019). Downstream behavioral and electrophysiological consequences of word prediction on recognition memory. Frontiers in human neuroscience, 13, 291. Retrieved from: https://www.frontiersin.org/articles/10.3389/fnhum.2019.00291/full

Jacoby, L. L., & Kelley, C. M. (1992). A process-dissociation framework for investigating unconscious influences: Freudian slips, projective tests, subliminal perception, and signal detection theory. Current Directions in Psychological Science, 1(6), 174-179.

Kane, M. J., Brown, L. H., McVay, J. C., Silvia, P. J., Myin-Germeys, I., & Kwapil, T. R. (2007). For whom the mind wanders, and when: An experience-sampling study of working memory and executive control in daily life. Psychological science, 18(7), 614-621.

Miyake, A., & Friedman, N. P. (2012). The nature and organization of individual differences in executive functions: Four general conclusions. Current directions in psychological science, 21(1), 8-14.

Stern, C. E., & Hasselmo, M. E. (2009). Recognition Memory. Retrieved from: https://www.sciencedirect.com/topics/agricultural-and-biological-sciences/recognition-memory

Related Samples

WeCreativez WhatsApp Support
Our customer support team is here to answer your questions. Ask us anything!
👋 Hi, how can I help?