Verb fluency task in individuals with mild cognitive impairment and dementia: a scoping review

Alves, Emily Viega; Só e Silva, Fernanda Mambrini; Souza, Victória Tizeli; Olchik, Maira Rozenfeld; Beber, Bárbara Costa; Castilhos, Raphael Machado

doi:10.1590/1980-5764-DN-2026-0476

person Emily Viega Alves

Conceptualization · Data curation · Formal analysis · Investigation · Methodology · Project administration · Validation · Visualization · Writing – original draft · Writing — review & editing

schoolUniversidade Federal do Rio Grande do Sul, Programa de Pós Graduação em Ciências Médicas, Porto Alegre RS, Brazil.

0000-0003-4800-3908

person Fernanda Mambrini Só e Silva

Data curation · Formal analysis · Investigation · Methodology

schoolUniversidade Federal do Rio Grande do Sul, Porto Alegre RS, Brazil.

0000-0003-3752-4169

person Victória Tizeli Souza

Data curation · Validation · Writing – original draft · Writing — review & editing

schoolUniversidade Federal do Rio Grande do Sul, Programa de Pós Graduação em Ciências Médicas, Porto Alegre RS, Brazil.

0009-0007-4143-2861

person Maira Rozenfeld Olchik

Validation · Writing – original draft · Writing — review & editing

schoolUniversidade Federal do Rio Grande do Sul, Programa de Pós Graduação em Ciências Médicas, Porto Alegre RS, Brazil.

0000-0002-8732-9225

person Bárbara Costa Beber

Validation · Writing – original draft · Writing — review & editing

schoolUniversidade Federal de Ciências da Saúde de Porto Alegre, Departamento de Fonoaudiologia, Porto Alegre RS, Brazil.

0000-0003-1781-3625

person Raphael Machado Castilhos

Conceptualization · Methodology · Software · Supervision · Validation · Visualization · Writing – original draft · Writing — review & editing

schoolUniversidade Federal do Rio Grande do Sul, Programa de Pós Graduação em Ciências Médicas, Porto Alegre RS, Brazil.

0000-0002-1905-2084

Correspondence: Emily Viega Alves; Email:
emiviegalves@gmail.com

This study was conducted by the Federal University of Rio Grande do Sul and the Hospital de Clínicas de Porto Alegre, Porto Alegre RS, Brazil.

Disclosure:

The authors report no conflicts of interest.

Editor-in-Chief:

Sonia M. D. Brucki. http://orcid.org/0000-0002-8303-6732

Associate Editor:

Renata Kochhann. https://orcid.org/0000-0002-6328-8131

Reference	Objective	Sample	VF assessment methodology	Main results	Conclusion
Piatt et al.¹⁷	To compare lexical, semantic, and action fluency in patients with PD and without dementia.	PD without dementia: mean age=70.3 years (SD=6.64). PD with dementia: mean age=73.9 years (SD=6.36). Controls: mean age=72.86 years (SD=7.51). Total sex distribution: 75 men and 61 women.	Duration: 1 minute Methodology: Analysis of correctly produced verbs	VF was significantly worse in patients with PD and dementia compared with those without dementia and controls. The task was more sensitive in differentiating groups than lexical and semantic fluency.	The VF task may be useful for detecting dementia in the early stages in patients with PD.
Ostberg et al.¹⁸	To compare noun, verb, and phonological fluency in individuals with varying levels of cognitive impairment.	AD: mean age=72.0 years (SD=10.0). MCI: mean age=66.7 years (SD=10.9). SCD: mean age=66.8 years (SD=11.2).	Duration: 1 minute (6 intervals of 10 seconds) Methodology: Analysis of correctly produced verbs	VF was disproportionately impaired in individuals with MCI, showing a declining pattern over time. Additionally, VF demonstrated greater discriminatory power among the types of fluency evaluated.	Impaired verb fluency may serve as a marker for the diagnosis of MCI.
Ostberg et al.¹⁹	To investigate the relationship between verb and noun fluency and cerebral perfusion deficits in individuals with different degrees of cognitive impairment.	AD (n=33): mean age=64.8 years (SD=9.4). MCI (n=33): mean age=60.5 years (SD=8.7). SCD (n=33): mean age=56.9 years (SD=7.4).	Duration: 1 minute (6 intervals of 10 seconds) Methodology: Analysis of correctly produced verbs	Individuals with AD produced fewer verbs in the VF task compared with the MCI and SCD groups. Temporal hypoperfusion and lower educational level were predictors of worse performance.	The task differentiated the evaluated groups, with worse performance in AD. Impaired VF may be associated with temporal lobe dysfunction.
McDowd et al.²⁰	To understand components of verbal fluency that differentiate groups with AD and PD, focusing on semantic and executive processes.	Young adults: mean age=21.5 years (SD=3.1). Older controls: mean age=72.0 years (SD=5.4). AD: mean age=73.8 years (SD=7.2). PD: mean age=71.9 years (SD=6.0).	Duration: Traditional format: 3 minutes; Alternating format: 6 minutes Methodology: Analysis of correctly produced verbs, perseverations, intrusions, clusters (number and size), switches, response time, and comparison between the two formats.	Young adults had better overall performance, while older adults with AD showed more intrusions, perseverations, and worse verbal fluency. Both the AD and PD groups showed fewer clusters, smaller clusters, and fewer switches. However, the PD group performed worse than the AD group. Individuals with AD and PD responded more slowly initially, with AD showing greater difficulty over time. The alternating format increased cognitive demand, resulting in greater impairment in patients, whereas the traditional format benefited all groups.	VF reveals semantic and executive differences between AD and PD, with processing speed having a greater impact on performance, especially in formats requiring cognitive switching.
Lai and Lin²¹	To investigate action-object semantic impairment in Chinesespeaking individuals with and without AD using naming fluency, VF, and picture naming tasks.	AD: mean age=77.55 years (SD=6.79); 7 women and 13 men. Adult controls: mean age=71.75 years (SD=5.93); 10 women and 10 men. Older controls: mean age=50.30 years (SD=4.74); 10 women and 10 men.	Duration: 1 minute Methodology: Analysis of correctly produced verbs	Patients with AD produced significantly fewer verbs compared with both adult and older controls, with age and disease being the main factors affecting task performance.	VF was a predictor of dementia severity in Chinese-speaking individuals.
Delbeuck et al.²²	To determine whether performance on the VF task has diagnostic value in differentiating dementia due to AD and DLB.	AD: mean age=71.64 years (SD=9.53); 14 men and 20 women. DLB: mean age=69.26 years (SD=10.02); 11 men and 7 women.	Duration: 1 minute Methodology: Analysis of correctly produced verbs	Individuals with AD performed better on VF compared with those with DLB. Noun fluency was also lower in the DLB group.	VF differentiates individuals with AD from those with DLB, with worse performance observed in DLB. Combining VF with noun fluency increases diagnostic accuracy.
Clark et al.²³	To evaluate noun, verb, and letter fluency in MCI and AD using lexical similarity measures and gray matter analysis of regions of interest.	AD: mean age=74.7 years (SD=7.8); 7 men and 3 women. MCI: mean age=70.7 years (SD=7.4); 18 men and 5 women. Controls: mean age=70.1 years (SD=6.9); 13 men and 12 women.	Not reported	Controls outperformed the MCI and AD groups in verb fluency. Semantic similarity strongly influenced word proximity in the task. VF was associated with activation of the left dorsal frontal cortex and temporal lobes.	Individuals with AD and MCI show deficits in verb fluency, related to dysfunction in frontotemporal regions.
Beber et al.²⁴	To investigate the nature of verb impairment in patients with mild and moderate AD by comparing verb fluency and verb naming tasks.	AD: mean age=77.74 years (SD=5.93); 22 women and 13 men. Controls: mean age=73.14 years (SD=4.82); 22 women and 13 men.	Duration: 1 minute Methodology: Analysis of correctly produced verbs	Patients with AD showed deficits in both naming and VF tasks, with greater impairment observed in naming. Naming performance declined with AD severity, while VF remained relatively stable.	Verb deficits in AD are predominantly semantic, with greater impairment in naming than in VF.
Mirandez et al.²⁵	To evaluate which verbal fluency task best discriminates between MCI and healthy individuals, analyze whether partial scores improve this discrimination, and investigate the correlation with CSF biomarkers for AD in the MCI group.	MCI: mean age=74.5 years (SD=5.4); 22 women and 8 men. Controls: mean age=72.5 years (SD=4.5); 26 women and 11 men.	Duration: 1 minute Methodology: Analysis of correctly produced verbs	Individuals with MCI produced significantly fewer verbs, especially in the second half of the VF task. There was no significant correlation between performance and CSF biomarkers, possibly due to the small sample size.	VF discriminated between MCI and healthy individuals. However, animal fluency alone and combined with letter fluency was slightly superior in discriminating controls from individuals with MCI.
Alegret et al.¹⁵	To investigate VF deficits in individuals with MCI and AD, assess the task’s potential for tracking disease progression, and propose normative scores for the Spanish population.	AD: mean age=79.0 years (SD=7.6); 235 women and 132 men. MCI: mean age=71.7 years (SD=8.9); 526 women and 369 men. Controls: mean age=63.3 years (SD=7.8); 391 women and 177 men.	Duration: 1 minute Methodology: Analysis of correctly produced verbs	Verb fluency performance declined significantly from healthy controls to MCI and AD, with lower scores associated with older age and lower educational level. In a longitudinal follow-up (mean duration of 16.6 to 19.3 months), verb fluency predicted conversion from healthy cognition to MCI (18.2%) and from MCI to dementia (21.6%).	VF was a sensitive and predictive marker for conversion from healthy cognitive status to MCI and dementia.
Macoir et al.²⁶	To evaluate the contribution of word production, especially verbs compared with nouns, for detecting cognitive impairment.	MCI: mean age=71.05 years (SD=6.1); 9 men and 11 women. SCD: mean age=66.4 years (SD=6.0); 4 men and 16 women. Controls: mean age=70.8 years (SD=7.1); 6 men and 14 women.	Duration: 1 minute Methodology: Analysis of correctly produced verbs, repeated words, errors, clustering and switching analyses (cluster size and number of switches), and word count across three time intervals (1–20s, 21–40s, 41–60s).	Verb production was lower in the MCI and SCD groups compared to controls, with no significant differences in errors or repetitions. Cluster size and number of switches were similar between controls and individuals with SCD. All groups showed a decline in verbal production over time.	VF is sensitive in detecting cognitive impairment in SCD, supporting its role as an early marker in the progression from healthy aging to MCI and dementia.
Paek et al.²⁷	To compare brain activation patterns between individuals with dementia and healthy older adults to identify regions associated with VF.	Dementia: mean age=68.11 years (SD=5.78); 2 men and 5 women. Controls: mean age=70.57 years (SD=7.55); 3 men and 6 women.	Duration: 90 seconds (3 intervals of 30 seconds each, with 10-second breaks in between) Methodology: Analysis of correctly produced verbs	Controls produced more verbs in intervals 2 and 3. Patients with dementia showed greater bilateral frontal activation. Medial temporal lobe activation was associated with worse VF performance.	Decline in VF task performance can be used not only as a cognitive marker of dementia but also as a neural marker of frontal and temporal lobe deficits observed in neurodegenerative diseases.
Paek²⁸	To examine whether individuals with amnestic AD produce verbs with higher emotional valence in VF responses compared with cognitively healthy individuals.	AD: mean age=78.6 years (SD=8.79); 5 men and 10 women. Controls: mean age=73.0 years (SD=7.08); 7 men and 10 women.	Duration: 30 seconds Methodology: Analysis of correctly produced verbs	Individuals with amnestic AD produced more verbs with higher emotional valence (more positive words) in VF compared with cognitively healthy individuals.	The higher emotional valence observed in individuals with AD may reflect emotional compensation associated with verbal memory impairment.
Scheffel et al.²⁹	To verify whether verbal fluency differentiates early variants of PPA and PAS before symptom overlap.	Logopenic PPA: mean age=66.3 years (SD=8.5); 29 women and 28 men. Semantic PPA: mean age=66.8 years (SD=6.7); 9 women and 6 men. Nonfluent PPA: mean age=68.5 years (SD=9.5); 15 women and 19 men. PAS: mean age=70.3 years (SD=9.2); 17 women and 13 men.	Duration: 1 minute Methodology: Analysis of correctly produced verbs	All groups showed abnormal performance on VF, but the task did not distinguish among PPA variants. Individuals with PAS performed better. VF showed a moderate correlation with aphasia severity.	VF does not differentiate among early variants of PPA, but performance is moderately associated with aphasia severity.
Beber et al.³⁰	To investigate VF performance in individuals with AD and healthy older adults, focusing on measures such as total words, clusters, and switches.	AD: mean age=78.6 years (SD=6.7); 20 women and 9 men. Controls: mean age=71.5 years (SD=6.3); 34 women and 5 men.	Duration: 1 minute Methodology: Analysis of correctly produced verbs, clustering and switching analyses (number of clusters, cluster size, and number of switches)	Individuals with AD produced fewer switches and fewer correct verbs compared to healthy older adults. There were no significant differences in the number or size of clusters between groups.	VF performance in AD predominantly reflects executive deficits, as evidenced by the reduced number of switches.
Fisher et al.³¹	To investigate the reliability of subjective clustering methods and the feasibility of an objective clustering method in the analysis of VF in AD and controls.	AD: mean age=78.15 years (SD=8.02); 15 women and 5 men. Controls: mean age=72.27 years (SD=6.94); 14 women and 8 men.	Duration: 1 minute (if the participant understands the command immediately) or 30 seconds (if there are two incorrect responses or initial non-comprehension). Methodology: Analysis of clusters and switches (total number of words, number of clusters, cluster size, and number of switches), combining subjective and objective analyses based on the Lancaster Norms database.	Individuals with AD produced significantly fewer words, clusters, and switches than controls, regardless of method (subjective or objective). Cluster size did not differ significantly between groups.	Objective clustering methods, such as those based on the Lancaster Norms, are feasible for analyzing VF in AD, showing agreement with subjective methods. Subjective analysis may help reduce bias and increase replicability.

[T1FN1] Abbreviations: VF, Verb Fluency; PD, Parkinson’s Disease; SD, Standard Deviation; AD, Alzheimer Disease; MCI, Mild Cognitive Impairment; SCD, Subjective Cognitive Decline; DLB, Dementia with Lewy Bodies; CSF, Cerebrospinal Fluid; PPA, Primary Progressive Aphasia; PAS, Progressive Apraxia of Speech.

[T1FN2] Source: The authors.

Verb fluency task in individuals with mild cognitive impairment and dementia: a scoping review

Tarefa de fluência de verbos em indivíduos com comprometimento cognitivo leve e demência: uma revisão de escopo