Order Recall in Verbal Short-Term Memory: The Role of Semantic Networks

Poirier, M., Saint-Aubin, J., Mair, A., Tehan, G., & Tolan, A. (2015). Order recall in verbal short-term memory: The role of semantic networks. Memory and Cognition, 43(3), 489–499. https://doi.org/10.3758/s13421-014-0470-6

Let’s say you were asked to recall the word list, “cheese, steak, wine,”. Because you know all these words, your brain would likely process both the phonological (speech sound) form and the meaning of the word. But how would this unfold? Would you first activate the phonological representation of each and then you draw on semantic knowledge (meaning-based information)? Or would both types of information be activated at the same time?

Verbal short-term memory has often been viewed as operating in separate stores. That is, the phonological form is held in mind and rehearsed using what is called, ‘the phonological loop’. At the point of recall, if the phonological trace has been degraded, then semantic information stored in long-term memory is activated to help with reconstruction of the word, a process referred to as redintegration. According to this view, semantic knowledge held in long-term memory is activated and transferred to short-term memory at the point of recall, and only after phonological information is processed. Recently, this 2-step process has been called into question.

According to the language-based model, verbal short-term memory reflects the direct (and automatic) activation of your phonological, lexical, and semantic representations for each word. There is considerable overlap between verbal short-term memory and language processing from the moment of encounter, and not just at the point of recall. If an integrated network activates all linguistic knowledge related to a word on encounter, then semantic effects should be observed to influence one of the most well-described short-term memory findings: serial recall, the immediate recall of items lists in correct order.

Poirier et al. (2015) completed a series of studies manipulating the semantic relationship of words in the list to a particular target word. In Experiment 1, participants were presented with a 6-item word list wherein the first three items were or were not semantically related to the target word (item 5). Recall of the stimuli in Experiment 1 could have been influenced by a potential grouping strategy, that is, the participants could have grouped semantically related words in the condition where the words were related. Therefore, in Experiment 2, all items were from the same semantic category and were either studied in silence (2a) or under articulatory suppression (2b). Participants were asked to recall the word list in the order presented across all experiments.

Across the three studies, the target word migrated towards earlier positions at the point of recall when semantically related words were present, and this was not the result of a grouping strategy. The authors suggest that the migration occurred as a result of coding serial position (i.e., order information) as part of the semantic network that supports language representations. This provides a new perspective on the processes involved in verbal short-term memory: language processing in short-term memory is closely tied to semantic networks in long-term memory, and thereby, short-term and long-term memory may be more integrated than they are distinct. This may have important implications for word and language learning such as thinking about how phonological rehearsal and semantic processing should be integrated during the learning phase, rather than learned discretely, in order to optimize on retention.

Blogger: Theresa Pham is a student in the combined SLP MClSc/PhD program, supervised by Dr. Lisa Archibald.

Morphological Awareness and Word-Level Reading in Early and Middle Elementary School Years

Robertson, E.K. & Deacon, S.H. (2019). Morphological awareness and word-level reading in early and middle elementary school years. Applied Psycholinguistics, 40, 1051-1071.

Much of previous research on reading development has examined the contributions of phonological awareness, but the role of morphological awareness is not as well documented. Morphological awareness is defined as one’s ability to identify and manipulate units of meaning in language, called morphemes. The small body of previous literature examining morphological awareness and reading in elementary school years has yielded mixed findings, with some evidence that its role in reading increases over time, some evidence that its role is relatively stable, and some evidence that its role declines. The present study took a cross-sectional approach to examine how the relationship between morphological awareness, word reading, and pseudoword reading may change in early and middle elementary school years.

Participants in the present study were 375 children in Grades 1 to 4, grouped into two groups for analysis purposes: one group of students in Grades 1 and 2, and a second group of students in Grades 3 and 4. Participants all completed a real word reading task, a pseudoword reading task, and a morphological awareness task which assessed production of the past-tense inflectional morpheme -ed. In this task, participants listened to a sentence and then were asked to finish a second sentence, for example: “We play games. Yesterday, we did the same thing; we ____ (played) games.” To ensure the morphological awareness task was sufficiently difficult for the older participants, it also included pseudoword items, such as “Jill can blick. Yesterday, she did the same thing; she ____ (blicked).” Participants also completed measures of phonological awareness, phonological short-term memory, sentence-level oral language skills, and nonverbal cognition, to be used as control measures in the analyses.

The authors used hierarchical regression analyses to examine the unique contributions of morphological awareness to real word reading and to pseudoword reading. With respect to real word reading, morphological awareness accounted for 1.1% of additional unique variance beyond the control measures in the Grades 1 and 2 group, however, it did not significantly account for additional variance in the Grades 3 and 4 group. With respect to pseudoword reading, morphological awareness explained an additional 1% beyond what was accounted for by the control variables, and no interaction was found with grade group.

These findings suggest that the role of morphological awareness in real word reading declines from early elementary school years to middle elementary school years, whereas its role in pseudoword reading remains stable over these grades. The authors suggest that the reduced contributions of morphological awareness in middle elementary school years maybe be related to the increased role of sentence-level language skills when reading, as children shift from learning to read to reading to learn and are exposed to texts that are more rich in morphosyntactic complexity and vocabulary. These findings suggest that morphology may be an important component of early reading instruction in the classroom and reading intervention, although further research on instruction and intervention approaches is needed to better understand how this might benefit reading outcomes.

Blogger: Alex Cross is an M.Cl.Sc. and Ph.D. Candidate in Speech-Language Pathology, supervised by Dr. Lisa Archibald and Dr. Marc Joanisse.

Executive Functioning and Narrative Language in Children with Dyslexia

Fisher, E. L., Barton-Hulsey, A., Walter, C., Sevcik, R. A., & Morris, R. (2019). Executive Functioning and Narrative Language in Children with Dyslexia. American Journal of Speech-Language Pathology, 28, 1127-1138.


We can think about language skills in terms of its structure, the vocabulary and grammar used to make sentences and send messages. We can also think about language in terms of its function, that is, the meaning of words (semantics) or stories (narratives) in our messages. There are many cognitive processes that support language structure and function. Phonological processing refers to the ability to use speech sounds to support oral and written language, and it supports structural aspects of language. Children with a specific reading disability known as dyslexia have been found to have poor phonological processing, and in particular, poor phonological awareness, or the awareness of understanding the sound structure of one’s language, and phonological recoding, or the ability to map sounds onto the letters of a language. Executive function is a term used to describe the set of skills that allow individuals to establish goal-oriented behaviours such as initiating, planning, and organizing.  Executive functioning consists of three skills: shifting, inhibition and working memory updating, and they support functional aspects of language ability. Some children with dyslexia have been found to have impairments in executive function as well.

The present study examined the influence of executive function on language abilities at the structural and functional level in a sample of children with dyslexia. Children with dyslexia completed measures of language structure and function, and measures of executive function. Results revealed a correlation between measures of inhibition and structural language abilities, and between measures of working memory and both structural and functional language. When examining groups based on both language and executive function, children with both low language and low executive function were found to have more difficulty with narrative language than the other groups. Additional analyses suggested that the working memory measure accounted for variance in the narrative language measures beyond what was accounted for by structural language.

These results suggest executive function skills support a child’s language ability. Specifically, working memory contributes to a child’s narrative retell abilities. This research adds to the evidence that there are a number of factors that support language ability and conducting assessments that look at multiple components of language and executive function will help a child’s language profile.

Blogger: Meghan Vollebregt is a student in the combined SLP MClSc/PhD program working under the supervision of Dr. Lisa Archibald.

Working memory profiles of children with dyslexia, developmental language disorder, or both

Gray et al. (2019). Working memory profiles of children with dyslexia, developmental language disorder, or both. JSLHR, 62, 1839-1858.

The ability to briefly hold information in mind and manipulate it is referred to as Working Memory. Working memory supports the completion of complex cognitive tasks in the moment. As we complete a task, we often have to remember some piece of information and add to it or change it as we go. Working memory supports this kind of work. It has been suggested that working memory supports both language learning and reading skills. Evidence showing an association with working memory has been demonstrated for both those with deficits in language learning known as Developmental Language Disorder (DLD) and those with dyslexia, a reading disorder usually attributed to a phonological processing deficit. Nevertheless, not all children with these disorders have been found to have a working memory impairment.

The purpose of this study was to systematically explore the relationship between working memory functioning and profiles of DLD and dyslexia. Children in grade 2 who either had DLD, dyslexia, both DLD and dyslexia, or typical development completed standardized tests of language, vocabulary, word reading, nonverbal intelligence, and others. The participants also completed a battery of working memory measures tapping storage and manipulation of information with auditory nonverbal, visual nonverbal, and number-based stimuli. The working memory battery also included storage-only tasks with either verbal or nonverbal stimuli, and tasks requiring binding of material that was either phonological or visuo-spatial, or both. Findings revealed that working memory profiles were not consistently associated with diagnostic category. That is, low (and high) working memory profiles occurred in children in the typically developing group, as well as in the DLD, dyslexia, and DLD/dyslexia group.

The authors argued that assessing working memory could contribute to the understanding of an individual child’s profile of strengths and weaknesses, which could help to direct intervention.

Blogger: Lisa Archibald

Rehearsal Effects in Adult Word Learning

Kaushanskaya, M., & Yoo, J. (2011). Rehearsal effects in adult word learning. Language and Cognitive Processes, 26(1), 121-148. doi:10.1080/01690965.2010.486579

Imagine you were trying to learn some useful Spanish words before travelling to Spain. You might start rehearsing the phrase ‘buenos días’ over and over again. This is one type of strategy people automatically use when learning new (foreign) words. By rehearsing the words, the novel phonological word form (speech sounds) is kept active in our minds using what’s called, ‘working memory’. Eventually, the word form is encoded (entered and stored) in long-term memory. There are two ways we might rehearse a word: (1) subvocal rehearsal (i.e., silent), that is just saying it in your head but not aloud and not moving your mouth, and (2) vocal rehearsal, that is saying the word out loud.  One of the goals of this study was to better understand how each of these processes support the learning of new words. Further, foreign words may contain sounds and/or sound structures that do not occur in English, such as /x/ in the Spanish word ‘ojo’ (meaning ‘eye’). Another goal of the study was to investigate how English-like words would benefit from rehearsal differently than words that contain non-native sounds. The also study wanted to examine how meaning is associated with learning novel words. This is important because learning a new word involves knowing both the phonological form and its associated meaning. Returning to the ‘buenos días’ example, it might be helpful to know that this means ‘good morning’ in English so that you can use it to greet people in the morning and not at night.

Across two experiments, adults who spoke only English learned novel words that followed the English structure (phonologically familiar) in Experiment 1 and novel words that contained non-native sounds (phonologically unfamiliar) in Experiment 2. Each word was paired with an English translation. Within each experiment, half of the words were learned through vocal rehearsal and the other half through subvocal rehearsal. Learning was assessed indirectly, that is, participants would hear the novel word and had to recall the English translation (recall task) or choose the correct English translation from five alternative words (recognition task). Testing was done immediately and one week later.

The results indicate that vocal rehearsal was beneficial for phonologically familiar words while subvocal rehearsal was beneficial for phonologically unfamiliar words when tested immediately. Interestingly, there was no difference at delayed testing, with weak rehearsal effects at minimum. The authors conclude that different rehearsal strategies may support learning in different ways. When saying aloud novel phonologically familiar words, this may highlight similarities between structures of novel words and English, and thereby, increase the learner’s ability to rely on native-language knowledge during learning. In contrast, saying aloud novel phonologically unfamiliar words may highlight differences and deter learning. Instead, silent rehearsal of phonologically unfamiliar words might be beneficial as it may not make such differences salient, and thereby, increase reliance on native-language knowledge. Findings from this work serves as an important first step into understanding mnemonic strategies that can potentially lead to better word learning and help create stronger links with long-term memory. Results must be interpreted with caution, however, given that the effects were observed for immediate but not long-term recall.

Blogger: Theresa Pham is a student in the combined SLP MClSc/PhD program, supervised by Dr. Lisa Archibald.

Learning and Overnight Retention in Declarative Memory in Specific Language Impairment

Lukács, Á., Kemény, F., Lum, J. A., & Ullman, M. T. (2017). Learning and overnight retention in declarative memory in specific language impairment. PloS one, 12(1), e0169474.

Memory is divided into a long-term memory system and a working memory system. Research has demonstrated that children with developmental language disorder (DLD; also referred to as specific language impairment) have poor working memory which means they have difficulty holding and manipulating information in mind that they have just received. There has been less research examining the long-term memory system in children with DLD. The long-term memory system has two types of memory; procedural and declarative memory.  Procedural long-term memory is our knowledge of how to do something such as ride a bike or tie our shoes and declarative memory is a type of long-term memory used for recalling facts, knowledge, events, and words.
The present study examined the role of declarative memory in children with DLD. The researchers assessed both immediate learning (10-minute delay) and retention (1-day delay) using nonverbal stimuli (seeing pictures of objects) and verbal stimuli (hearing words). Children who were typically developing and children with DLD were asked to complete a recognition task that was used to assess declarative memory. The recognition task was a judgement task where the participants had to indicate if they had seen or heard the item before. Participants first completed an encoding phase where they were presented with 32 real and 32 novel items. This was completed for both nonverbal and verbal stimuli. They completed the judgement task 10-minutes after the encoding phase and again 1-day after the encoding phase.

The researchers found that for nonverbal stimuli the typically developing children showed no change in accuracy between the 10- minute delay and the 1-day delay but children with DLD were significantly more accurate at the 1-day delay compared to the 10-minute delay. The typically developing children performed significantly better than the children with DLD when tested after the 10-minute delay but there was no difference between groups at the 1-day delay. This result indicates that after some consolidation the children with DLD were more accurately able to identify nonverbal items seen before. For verbal items, typically developing children performed better than children with DLD and they did not find the same trend of better performance in the DLD group at the 1-day delay. This result was not surprising since children with DLD are known to have difficulty with word learning.

These results would suggest that children with DLD have consolidation strengths in declarative memory, specifically for non-verbal items. This finding demonstrates the importance of consolidation and sleep for learning new objects. It would suggest that time for consolidation might be important when determining what a child with DLD has learned.

Blogger: Meghan Vollebregt is a student in the combined SLP MClSc/PhD program working under the supervision of Dr. Lisa Archibald.

Mathematical thinking in children with developmental language disorder: the role of pattern skills and verbal working memory

Fyfe, E.R., Matz, L.E,, Hunt, K.M., & Alibali, M.W. (2019). Mathematical thinking in children with developmental language disorder: the role of pattern skills and verbal working memory. Journal of Communication Disorders, 77, 17-30.

This paper examines the role of pattern skills and verbal working memory in mathematics performance in children with developmental language disorder (DLD). Growing evidence suggests that children with DLD have difficulty on mathematics tasks such as counting and arithmetic (see Cross, Joanisse & Archibald, 2019, for a review). Working memory and pattern skills are both known to be associated with mathematical abilities. Some previous studies have identified difficulties with verbal working memory in children with DLD, as well as difficulties learning implicit patterns in statistical learning tasks. The aim of the study was to examine how working memory and pattern skills might contribute to differences in mathematics performance in children with DLD and typically developing (TD) children.

Participants were 36 children aged 6 to 13 years old, eighteen of whom had been identified with DLD. Children completed a battery of tasks assessing verbal working memory, pattern extension, math calculation, and knowledge of math concepts. Verbal working memory was assessed using the Competing Language Processing Task, in which children answered yes/no questions about sentences while also holding the last word of each sentence in memory. In the pattern extension task, children were given 7 elements which varied in shape and size and were asked to identify the 8th shape and explain their reasoning for selecting that shape. The math tasks consisted of math problems involving standard arithmetic (e.g. 2 + 4 + 5 + 2 = __ ), inversion ( e.g. 4 + 7 – 7 = __ ), and equivalence (e.g. 3 + 4 + 6 = 3 + __ ). For each math problem children were asked to explain their strategy, and the authors coded these responses based on whether they demonstrated conceptual understanding of inversion and equivalence.

The authors first used logistic regression to examine differences between the TD and DLD groups on each of the behavioural tasks. They found that, relative to the TD group, the group with DLD performed more poorly on each type of math problem and on the verbal working memory task, but performed similarly on the pattern extension task. A hierarchical regression was then used to examine how math calculation and knowledge of math concepts might each be predicted by pattern extension and working memory. Working memory accounted for some differences in calculation and concepts scores however it did not uniquely predict variance beyond the variance predicted by the group variable. Pattern extension performance did not predict concept scores but uniquely predicted calculation scores, above and beyond the variance predicted by group, age, non-verbal IQ, and working memory.

Overall, these findings suggest that some aspects of poor math performance in DLD may be related to verbal working memory capacity. Additionally, the ability to recognize patterns may support recognition of rules in number sequences, facilitating calculation skills in both TD children and children with DLD. Importantly, the authors acknowledge that their regression models accounted for less than half of the variance in calculation skills and concept knowledge, which suggest there any many other factors that may be contributing to development of mathematics knowledge. In general, these findings highlight that the linguistic and working memory-related factors in DLD can impact performance in many academic areas, despite other intact cognitive processes, including patterning.

Blogger: Alex Cross is an M.Cl.Sc. and Ph.D. Candidate in Speech-Language Pathology, supervised by Dr. Lisa Archibald and Dr. Marc Joanisse.

Describing language assessment for school-aged children: A Delphi study

Denman, D., Kim, J., Munro, N., Speyer, R., & Cordier, R. (2018). Describing language assessment for school-aged children: A Delphi study. International Journal of Speech-Language Pathology, 1-11.

The recent CATALISE studies reported a consensus for identification and terminology for children with persistent language difficulties with a significant functional or educational impact. A number of important questions follow on from the consensus statement. The purpose of this statement was to investigate the different types or purposes of language assessments completed by speech-language pathologists (SLPs).

Informed by current evidence and conversation, the researchers developed a taxonomy for assessment that included (1) modality/domain, (2) purpose, (3) delivery, and (4) form. A group of 55 Australian SLPs agreed with this taxonomy on each of 2 rounds of a Delphi survey. In rounds 2 and 3 of the survey, participants were also asked to assign categories based on the taxonomy for 2 case studies. Although lower, agreement for these categorizations reached the established consensus target. SLPs often went beyond the details of the case to explain how one aspect of assessment could fit different categories.

The taxonomy reveals the multiple purposes and types of assessments completed by SLPs. SLPs often move flexibly across assessment purposes and types in order to address questions arising in an assessment.

Blogger: Lisa Archibald

The Role of Nonverbal Working Memory in Morphosyntactic Processing by Children with Specific Language Impairment and Autism Spectrum Disorders

Weismer, S. E., Davidson, M. M., Gangopadhyay, I., Sindberg, H., Roebuck, H. & Kaushanskaya, M. (2017). The role of nonverbal working memory in morphosyntactic processing by children with specific language impairment and autism spectrum disorders. Journal of Neurodevelopmental Disorders, 9(28). doi: 10.1186/s11689-017-9209-6.

The grammar of a language refers to the rules for how we put words and sentences together. One part of grammar is called morphology and refers to the smallest units of meaning in a language. Take, for example, the word ‘jumped’. Jumped has 2 meaningful units or morphemes – the action, jump, and the past tense marker that tells us the action is already over, -ed. Morphemes like ‘-ed’ are called inflectional morphemes, and there are a lot of them in the English language. Like ‘-ed’, some are bound to verbs to signal tense (when something is happening) or agreement (between the subject and verb of a sentence as in he jumps vs. they jump). Others mark nouns as plural (e.g., cups), possessive (e.g., Jill’s), etc. Children with a developmental language disorder (DLD; also known as specific language impairment) have difficulty learning these inflectional morphemes. Some studies also suggest that children with autism spectrum disorder (ASD) have the same kind of trouble with learning morphology.

Both children with DLD and ASD are also reported to have poor working memory. Working memory is a difficulty holding and manipulating information you’re thinking of in the moment. According to one theory, having poor working memory might make it difficult to learn language. If that’s the case, we would expect language tasks to be harder when more language is involved, for example, when sentences are long or complex. If a language error happens early in a sentence when there’s not much to hold in memory, it would be easy to detect the error even if working memory is poor. If a language error happens late in the sentence, however, someone with poor working memory would have a hard time remembering the whole sentence and catching the error. There’s another theory, though, that says that understanding the ends of sentences might be easier. According to the ‘wrap up’ theory, errors occurring later in a sentence will be more easily detected because there is more information or context available later in a sentence to help the child process and notice errors in grammar.

In this study, children who were typically developing or had DLD or ASD completed tests of working memory (for information unrelated to language) and grammar. The grammar test was a “grammatical judgement task” and involved the children listening to a sentence and deciding if that sentence contained any errors in grammar or not. The grammar errors were either close to the beginning or the end of the sentence. The researchers found that the typically developing, DLD and ASD groups did not differ on the test of working memory. On the grammatical judgement task, the researchers found that children with typically developing language skills were better at detecting grammar errors than children with DLD. This result is not surprising considering that children with DLD are known to have difficulty learning the inflectional morphemes that were used as the grammatical errors. As well, all children were better and faster at detecting grammatical errors late in a sentence than those occurring early in a sentence. This finding supports the “wrap-up” account. It suggests that the information in a sentence may help a child to make sense of that sentence and to notice any errors that may exist.

The results of this study suggest that context helps. That is, important details are best provided once a child understands the topic or context being discussed. It might suggest that the strategy of ‘Keep It Short and Simple’ may not be the best strategy for all children. Some children might benefit from having sufficient information to support their understanding.

Blogger: Alyssa Kuiack is a M.Cl.Sc./PhD candidate, in speech-language pathology, working under the supervision of Dr. Lisa Archibald.