Real Word and Non-Word Repetition Skills in Individuals With Broca’s Aphasia

INTRODUCTION

Broca’s aphasia, also known as expressive aphasia or non-fluent aphasia, is a type of language disorder resulting from damage to the posterior inferior frontal gyrus of the brain, known as Broca’s area. This damage can occur due to a stroke, traumatic brain injury, tumour, or infection. Individuals with Broca’s aphasia (IBAs) typically have difficulty with speech production, grammar, and complex language comprehension, but their understanding of simple sentences and their intelligence generally remain intact.^[1]

To assess language skills in IBAs, repetition tasks are often used as a diagnostic tool and a way to evaluate the extent of the impairment. These tasks minimise the cognitive demand for spontaneous speech generation and instead isolate specific language processing deficits.^[2] These tasks are used because they do not require the individual to generate a response but rather to replicate heard language, which can help to isolate the language processing components affected in individuals with aphasia. Real-word (RW) repetition tends to benefit from stored lexical and semantic knowledge, whereas non-word (NW) repetition, which relies on meaningless sound sequences, tests phonological processing and working memory without the support of lexical familiarity.^[3]

Repetition deficits in Broca’s aphasia arise not only from damage to Broca’s area but also from disruptions in regions within the language network.^[4] For RW repetition, individuals may be able to access known words from their lexicon but struggle with the motor planning and execution necessary for articulation. NW repetition, on the other hand, depends on the temporary storage and manipulation of phonological information, which is a function of working memory.^[5]

The dual-route model of repetition helps to explain the performance differences between RW and NW repetition tasks. According to this model, words are repeated by accessing their articulatory specifications directly from the lexicon. Because NWs have no lexical entry specifying an articulatory plan, they must be repeated by being segmented into their constituent units, which are then assembled into novel articulatory plans.^[6,7]

Repetition is a complex cognitive process that involves multiple cognitive skills, including auditory processing, phonological processing, articulation, and working memory. Studies have explored various types of repetition deficits in aphasia, including phonological, semantic, and verbal fluency deficits.^[8] Among these, NW repetition tasks have shown particularly strong sensitivity for identifying underlying phonological processing deficits, with impairments reported in up to 80% of individuals with aphasia.^[9]

Language characteristics also influence the complexity of repetition tasks. For example, Kannada, being a phonologically rich and agglutinative language, poses unique challenges for IBAs. Kannada’s complex phonotactic rules, consonant clusters, and morphological structure demand precise phoneme sequencing and coordination.^[10] IBAs often exhibit errors such as phoneme deletions and substitutions during repetition, reflecting deficits in phonological encoding and motor planning.^[11] RW repetition in Kannada may benefit from residual lexical knowledge, whereas NW repetition more directly exposes fundamental phonological processing difficulties.

Understanding the variability in repetition performance is essential. While some IBAs can repeat familiar words with moderate success, others show marked impairments, especially with NWs. This variability underscores the importance of examining both task types to understand the underlying deficits. RW repetition relies on lexical retrieval, while NW repetition relies on phonological working memory and processing.^[12]

A more nuanced understanding of repetition difficulties can be achieved by examining the relationship between accuracy and reaction time (RT). In IBAs, characterised by phonological and motor speech deficits, the interaction between these measures may provide insight into the efficiency of processing. A significant negative correlation would support the idea that individuals who perform more accurately also respond more quickly, while a lack of correlation or a positive relationship might indicate that accuracy and RT operate independently, possibly due to increased cognitive effort in speech planning.^[13] Additionally, investigating the correlation between RW and NW accuracy can determine whether performance on one type of task predicts performance on the other, which would highlight whether similar cognitive processes underlie both tasks.^[14]

The findings of the studies reporting the significant correlations between accuracy and RT vary. Some individuals with aphasia demonstrate fast and accurate processing, while others show no consistent relationship or exhibit slower yet accurate responses.^[15] In contrast, research on neurotypical controls found a positive correlation between accuracy and RT in NW repetition tasks, indicating that faster responses sometimes came with a trade-off in accuracy, possibly due to cognitive constraints.^[16] Additionally, another study investigated the relationship between RW and NW accuracy and found that performance on one task could predict performance on the other, suggesting that both tasks rely on similar cognitive processes, particularly phonological working memory and lexical access.^[17]

These studies highlight the need for further exploration of these correlations in individuals with aphasia to better understand the cognitive mechanisms underlying their language deficits and inform targeted interventions. Variations in findings on repetition impairments across studies highlight the diversity in aphasia presentations, types of tasks used, and sample characteristics, emphasising the need for further research. The present study aimed to investigate repetition impairments in IBAs by comparing performance on RW and NW repetition tasks. Additionally, we also examined the relationship between RW and NW repetition in terms of accuracy and RT among IBAs and NTIs.

MATERIAL AND METHODS

RESULTS

The tabulated data were subjected to both descriptive and inferential statistics using the Statistical Package for Social Sciences (SPSS) version 20.0. Descriptive statistics were done to find the mean and standard deviation obtained for RW and NW repetition tasks for subjects of both groups. The data was further subjected to a test of normality using the Shapiro-Wilk test, and the normality test revealed a non-normal distribution of the data (p > 0.05). Owing to these, non-parametric tests were used for data analysis. A Mann-Whitney U test was conducted to compare the performance between IBAs and NTIs on accuracy and RT for both RW and NW tasks, and within-group comparison was done using the Wilcoxon Signed-Rank test. Additionally, Pearson’s product-moment correlation analysis was performed to assess the relationship between IBA and NTI across four measures, including real-word accuracy (RW_Acc), real-word reaction time (RW_RT), non-word accuracy (NW_Acc), and non-word reaction time (NW_RT).

The mean values in Table 1 show that IBAs have poorer means compared with NTIs. Further, it is also clear that the accuracy scores were poorer for NW than RW repetition tasks in both groups.

The mean values in Table 2 showed that IBAs have greater means for RT than NTIs on both tasks. Further, it is also evident that the RT is greater among IBAs for the NW repetition task in comparison to the RW repetition task.

The results of the Mann-Whitney U test revealed that there is a significant difference in accuracy between the groups for both RW (U value of 225.00, Z = 4.92, and a p-value of 0.00) and NW (U value of 223.00, Z = 4.611, and a p-value of 0.00) repetition tasks. Similarly, RT measures also showed a significant difference between the groups, with respect to RW_RT (U value of 62.00, Z = 2.09, and a p-value of 0.036) and NW_RT (U value of 37.00, Z = 3.133, and a p-value of 0.002). The results of the Wilcoxon signed-rank test indicated that the accuracy scores (W= 57.500, p=0.05) differ significantly between RW and NW. At the same time, the RT measure did not show any significance (W= 262.000, p=0.336) for both RW and NW repetition tasks.

The results of Pearson’s product-moment correlation analysis have been illustrated in Table 3. For the IBA group, a strong positive correlation was observed between RW_Acc and NW_Acc (|r| = 0.843, p < 0.01), indicating that individuals who performed well on RW tasks also tend to perform well on NW tasks. This suggests a shared cognitive mechanism underlying accuracy in linguistic tasks for this group, potentially reflecting compensatory strategies or reliance on overlapping neural resources. However, no other significant correlations were found in the IBA group, indicating that RTs for RW and NW tasks did not show consistent linear relationships with accuracy measures.

In contrast, for the NTI group, correlations involving RW_Acc and NW_Acc could not be computed because the variables were constant, lacking the variability required for correlation analysis. This constant value may indicate uniformly high performance among NTI on these measures, which precludes meaningful statistical comparisons. Furthermore, no significant correlation was observed between RW_RT and NW_RT in the NTI group, suggesting that RTs for RW and NW tasks operate independently in NTI, reflecting distinct processing mechanisms for these measures.

Overall, these findings highlight a notable difference between the IBA and NTI groups in how linguistic tasks are processed. While IBAs exhibit a strong link between accuracy measures, likely due to overlapping or impaired neural systems, NTIs demonstrate independent mechanisms and consistent high accuracy that do not allow for variability-based analyses. These results underscore the unique ways in which aphasia impacts language processing and the compensatory strategies that IBAs may employ.

DISCUSSION

The relationship between phonological memory, lexical retrieval, and speech production is fundamental to understanding the cognitive deficits associated with IBA. Phonological memory, which involves the temporary storage and manipulation of phonological information, plays a crucial role in supporting both lexical retrieval and speech production. In IBA, damage to the left perisylvian region often results in impairments to phonological memory, disrupting the ability to store and reproduce word forms and causing difficulties in repetition tasks.^[24] Similarly, lexical retrieval, or the ability to access and retrieve stored word representations, is frequently affected due to damage to critical language areas, such as the inferior frontal gyrus.^[25] This impairment interferes with the mapping of lexical and phonological forms, leading to reduced speech fluency and accuracy.^[26]

The lexical status of a word plays a critical role in facilitating retrieval and articulation processes, which are often disrupted in IBA. Research has demonstrated that IBAs perform better when producing RWs compared to NWs, a phenomenon known as the “lexicality effect.”^[27] Real words benefit from pre-existing semantic and phonological representations in the mental lexicon, providing essential support for retrieval processes that are otherwise impaired. Neurological studies reveal greater activation in preserved language networks, such as the left superior temporal and inferior frontal regions, when processing RWs compared to NWs.^[28]

The observed differences in accuracy and RT between IBAs in RW and NW repetition tasks can be attributed to the fundamental linguistic and cognitive deficits associated with the condition. The significant differences in accuracy, as revealed by the Mann-Whitney U test, suggest that IBA struggles with both RW and NW repetition, with additional difficulty in NW tasks. This aligns with previous research indicating that while RWs benefit from lexical-semantic representations stored in the mental lexicon, NWs require greater reliance on phonological working memory and phonological assembly, processes that are typically impaired in IBAs.^[25,29]

The significant difference in RTs further supports the notion that IBA experiences delays in both lexical access and motor speech execution. Delayed processing in RW and NW tasks could stem from impairments in phonological encoding and articulatory planning, which are known to be compromised due to lesions affecting the left inferior frontal gyrus.^[30]

Within-group comparisons using the Wilcoxon signed-rank test revealed a significant difference in accuracy between RW and NW repetition, highlighting the additional challenge posed by NW tasks. This can be attributed to the fact that NW repetition is highly dependent on phonological working memory, which is often impaired in aphasia, making it harder to process and accurately repeat unfamiliar phonemic sequences.^[31] However, RT differences between RW and NW within IBAs were not statistically significant. This suggests that while accuracy declines for NW repetition, the cognitive and motor demands for producing RWs and NWs may be comparably high, leading to similarly prolonged RTs for both.^[32] These findings collectively emphasise the role of phonological and motor deficits in repetition tasks and support previous literature on the speech-processing challenges in Broca’s aphasia.

Correlation analysis offered additional insights into the relationship between accuracy and RT. In the IBA group, a strong positive correlation was observed between RW_Acc and NW_Acc, indicating that individuals who performed well on one task tended to perform well on the other. This suggests a shared cognitive mechanism for processing linguistic stimuli and may reflect compensatory strategies in IBA, where individuals rely on specific resources to enhance performance.^[33] Notably, no significant correlation between accuracy and RT was found in the IBA group, indicating a possible dissociation between these measures. This supports the notion that language production and comprehension deficits in IBA are not necessarily tied to processing speed, with some individuals exhibiting slow yet accurate responses.^[34]

The lack of significant correlations between RTs and accuracy measures in the IBA group is somewhat contradictory to previous findings, where such a relationship is typically observed. Previous studies have shown that RTs often correlate with accuracy in cognitive tasks, as both are considered markers of cognitive efficiency.^[35] However, the absence of such correlations in the IBA group may be attributed to individual variability in cognitive processing and language function within this population. The IBA group could exhibit a greater degree of cognitive heterogeneity, which might influence how RT and accuracy are related. Another study by Besson et al. suggests that individuals with specific aphasia subtypes or atypical brain lesions might not demonstrate the typical linear relationship between RT and accuracy, particularly when their cognitive or language-processing abilities are more severely impaired.^[36] Additionally, individuals in the IBA group might employ compensatory cognitive strategies such as heightened cognitive effort. This could result in longer RTs without affecting accuracy. It is also possible that the tasks used in this study were not sensitive enough to capture a correlation, or that neuroplasticity in the IBA group leads to atypical brain functioning that disrupts this relationship.^[37] Overall, the discrepancy may reflect the unique neural and cognitive profiles of the IBA group, underscoring the complexity of assessing RTs and accuracy in populations with atypical brain structures.

In contrast, correlations between RW and NW accuracy in NTIs could not be computed due to constant values, reflecting uniformly high performance. This lack of variability aligns with findings that NTIs exhibit robust and consistent performance across linguistic tasks, supported by intact phonological and lexical networks.^[27] Additionally, the lack of significant correlations between RW and NW RTs in NTIs suggests that distinct cognitive processes govern response timing for familiar versus unfamiliar stimuli, consistent with the modular view of language processing.^[38]

These results emphasise key differences in how IBAs and NTIs process linguistic tasks. For IBAs, the strong correlation between RW and NW accuracy highlights the reliance on shared or overlapping cognitive resources, potentially reflecting compensatory neural strategies following left-hemisphere damage.^[39] NTIs, on the other hand, demonstrate consistently high performance and independent processing mechanisms, suggesting distinct and robust neural pathways for different linguistic domains.

Overall, this research advances our understanding of the neural and cognitive underpinnings of BA and provides a foundation for developing targeted therapeutic approaches to address language deficits in affected individuals. Future research should explore targeted interventions, such as those focused on phonological working memory, and investigate their impact on speech outcomes to improve clinical strategies for treating aphasia.

CONCLUSION

In conclusion, the current findings highlight significant differences in RW and NW repetition skills between IBA and NTI, with IBAs showing marked deficits in both accuracy and reaction time. Further, correlation analysis revealed a strong positive correlation between RW and NW accuracy in the IBA group, suggesting a shared cognitive mechanism underlying both tasks. However, no such significant correlation was observed in the NTI group, indicating that different processing strategies may be involved. These findings underscore the cognitive and linguistic impairments inherent to aphasia, particularly in tasks requiring phonological processing and lexical retrieval. While IBAs demonstrate variability in performance due to impaired or overlapping neural mechanisms, NTIs exhibit consistent and efficient repetition abilities. The results emphasize the need for targeted therapeutic strategies that account for the unique linguistic challenges faced by individuals with Broca’s aphasia and support the development of compensatory mechanisms to improve language processing and communication outcomes.