Effects of one avocado a day for six months on cognitive performance in overweight adults: A randomized controlled trial

Abstract

Background

With the rising incidence of neurodegenerative diseases and aging of the global population, it is important to investigate foods that may preserve cognitive function with advancing years. Avocados are rich in fiber, carotenoids, antioxidants, polyphenols and monounsaturated fat, all of which have been associated with improved cognition.

Objective

To determine if consuming one avocado per day for 6 months has cognitive benefits in adults with central obesity, addressing cognitive health early in the aging trajectory.

Methods

This was an ancillary study of the multicenter Habitual Diet and Avocado Trial (HAT), which investigated the effect of avocado consumption on visceral adiposity. The effect of avocado consumption on cognition was assessed in participants at the Loma Linda University center. Participants with elevated waist circumference (n = 251) were randomly assigned to consume 1 avocado per day (AVO), or to avoid consuming avocados (CON) for 6 months without additional dietary guidance. Cognitive function (memory, processing speed, executive function, simple reaction time, and working memory reaction time) was measured at baseline and end of the intervention period. A series of mixed analyses of covariance was conducted to assess changes in the five cognitive scores over the 6-month intervention, group differences (AVO vs. CON), and modification of intervention effects by age.

Results

A total of 241 participants (69.7% female, 53.1% white, mean BMI = 33.4 ± 5.7) between the ages of 25–84 years (mean age = 49.6 ± 13.3 years) completed the trial. Working memory reaction time improved for all participants at the end of the 6-month intervention, but no significant group by time interactions were found between the AVO and CON groups in any of the cognitive domains. There were no significant two-way interactions with age by time or intervention group, and the three-way interaction was also non-significant.

Conclusions

The consumption of one avocado per day without any additional lifestyle modifications for six months did not significantly alter cognitive function in adults with central obesity across all age groups. Additional work is needed to determine whether avocados, as part of dietary strategies initiated in midlife, contribute to healthy cognitive aging, particularly in normal weight and metabolically vulnerable populations.

1. Introduction

The global population is aging [1], and the number of older adults diagnosed with neurodegenerative conditions such as Alzheimer’s disease (AD) continues to rise, along with other chronic diseases that adversely affect cognitive function [2]. Importantly, the biological and metabolic processes underlying cognitive aging often begin decades before the onset of clinically detectable impairment, with midlife representing a critical window for intervention [1].

Dietary patterns rich in antioxidants, unsaturated fats, and fiber have been associated with favorable effects on brain health and cognitive performance [3]. Polyphenol-rich foods such as blueberries and cocoa have been shown to enhance cerebral blood flow and brain activation [4] and are linked to improved cognitive performance and reduced cognitive decline in older adults [[5], [6], [7], [8]]. Nuts, rich in unsaturated fats, fiber and bioactive phytonutrients have been associated with better cognitive outcomes and slower rate of cognitive decline across longitudinal studies [9,10].

The avocado is a unique fruit that provides a combination of monounsaturated fatty acids (MUFA), fiber, and antioxidant phytochemicals such as polyphenols and carotenoids [11]. These dietary components have been implicated in key pathways relevant to cognitive aging, including vascular function, oxidative stress, inflammation, and insulin sensitivity. Diets abundant in MUFA and fiber have been associated with improved global cognition and protection against age-related cognitive decline [[12], [13], [14]]. The mesocarp of the avocado is particularly rich in the carotenoids xanthophyll, lutein and zeaxanthin, which have been linked to enhanced cognitive efficiency, memory, and neural integrity [[15], [16], [17], [18]].

To date, only two clinical trials have examined the effects of avocado consumption on cognition, with each reporting limited benefits in select aspects of executive function [19,20]. However, only one of these studies assessed multiple cognitive domains and was conducted in a healthy, normal-weight older adult population [19] while the other was conducted in adults with overweight and obesity, but only in young to middle-age adults (age 25–45 years), and only focusing on a single cognitive domain (executive function) [20]. The present study extends this evidence base by evaluating the effects of consuming one avocado daily for six months on multiple domains of cognitive performance in adults from young adulthood to late life, to examine potential age-related influences on the effect of avocados, and with central obesity, a group at elevated risk for metabolic dysfunction and cognitive decline.

2. Methods

The present study was an ancillary study of The Habitual Diet and Avocado Trial (HAT), which was designed to determine if consuming one avocado per day as part of one’s habitual diet for 6 months could elicit measurable changes in visceral adiposity in a cohort of adults with central obesity [21]. This multicenter, parallel, randomized clinical trial (RCT) was conducted by research centers at Loma Linda University (LLU), the University of California, Los Angeles, Tufts University, and Pennsylvania State University (University Park and Hershey campuses), with Wake Forest University serving as the coordinating site. The trial is registered at https://clinicaltrials.gov (NCT 03528031). Changes in cognition were assessed as part of an ancillary study conducted with participants at the LLU site only. The Institutional Review Boards of Wake Forest University and Loma Linda University approved of this research and all participants provided written informed consent. A description of the methods implemented for the parent HAT study have been published [22].

2.1. Participants

Participants were recruited via email, social media, news outlets, and word of mouth. Those interested were invited to attend an information session where eligibility was assessed in one-on-one interviews between research clinicians and potential participants. This multistage screening and selection process has been used in previous trials [10]. Participants were required to have an elevated waist circumference (≥35 inches for women, ≥40 inches for men), ≥25 years of age at screening, and consume no more than 2 avocados per month. Exclusion criteria included any allergies/sensitivities to avocados, a diagnosed neurological or cognitive disorder, unstable medical conditions, and pregnancy/lactation or intention to become pregnant. Of the 1005 participants selected across the 4 study centers, 251 were enrolled at the LLU site for the primary study and also recruited for this ancillary study.

2.2. Study design/intervention

Participants were randomized to either the avocado group (AVO) or the control group (CON). The AVO group was provided with fresh Hass avocados and instructed to consume one avocado per day without instructions about changes to their habitual diet, while the CON group was instructed to simply continue with their habitual diet and to limit their avocado intake to ≤2 avocados per month. Those randomized to the AVO arm received a recipe booklet as well as a brief consultation with research dietitians for guidance on how to incorporate the avocado into their daily diets. Both groups were asked not to make any changes to their lifestyle during the study. Any deviations from the protocol were required to be documented in a lifestyle journal. Participants in the AVO group came to the nutrition research center (NRC) to pick-up avocados once every two weeks, and all participants had a clinic visit with the study dietitian at baseline, and every month until the final visit at the end of the intervention. Height and weight were measured at each visit and Body Mass Index (BMI) was calculated.

2.3. Dietary assessment and compliance

24-h diet recalls were conducted via telephone by research dietitians at baseline, 2 months, 4 months, and 6 months to estimate intake prior to the study as well as over the course of the study. Dietary data were recorded and analyzed using the University of Minnesota Nutrition Data System for Research (NDSR, version 2018). Participants were provided with a ‘Food Amounts Booklet’ for reference and recalls were unannounced to minimize reactivity.

2.4. Assessment of cognition

Cognition was assessed prior to randomization (baseline) and at the end of the 6-month intervention (Table 1). The test battery included several paper-and-pencil tests, including the Rey Auditory Verbal Learning Test (RAVLT), Brief Visuospatial Memory Test – Revised (BVMT-R), Symbol Digit Modalities Test (SDMT), Trail Making Test (TMT), Stroop Test (Golden version), FAS, Digit Span, and Auditory Consonant Trigrams (ACT). Additionally, the following tasks from the Cogstate computerized battery [32] were administered: Detection Test, Identification Test, One Back Test, and Two Back Test. The test battery was administered by trained psychometrists. Participants were evaluated by the same psychometrist at each visit, at the same time of day, and were instructed to follow a similar routine before each visit in terms of amount of sleep, coffee consumption, and other daily activities.

Table 1.

Cognitive battery.

Composite score	Cognitive tests
Memory	RAVLT Total Immediate Recall
	RAVLT Delayed Recall
	BVMT-R Total Immediate Recall
	BVMT-R Delayed Recall
Processing speed	Symbol Digit Modalities Test
	Trail-Making Test, Part A
	Stroop Word Reading
	Stroop Color Naming
Executive function	Digit Span
	Auditory Consonant Trigrams
	Trail-Making Test, Part B
	Stroop Color-Word Interference
	FAS
Simple reaction time	Cogstate Detection Test
	Cogstate Identification Test
Working memory reaction time	Cogstate One-Back Test
	Cogstate Two-Back Test

Note. RAVLT = Rey Auditory Verbal Learning Test; BVMT-R = Brief Visuospatial Memory Test - Revised.

Composite scores: Three composite scores were calculated based on the paper-and-pencil tests, and two composite scores were calculated based on the mean reaction time scores from the Cogstate. Within each cognitive domain, all relevant subtest scores were standardized using proportion of maximum scaling scores before calculating mean scores [23]. This procedure transforms each scale to a metric from 0 (minimum possible) to 1 (maximum possible). For ease of interpretation, scores were multiplied by 100 to yield the percentage of the maximum possible score. Scores that take the form of completion times (TMT Parts A and B) or reaction times (Cogstate Detection, Identification, One-Back and Two-Back Tests) were reversed during standardization so that higher scores reflect better performance.

2.5. Statistical analysis

To evaluate whether our five cognitive outcomes (memory, processing speed, executive function, simple reaction time, and working memory reaction time) differed by intervention group (AVO vs. CON) and study time point (baseline vs. end of intervention) and whether there was an interaction between intervention and time, we ran five separate 2 (Time) × 2 (Intervention Group) mixed analyses of covariance (ANCOVAs) controlling for participant age at baseline. We also included the three-way interaction between our covariate (age), time, and intervention group, as well as all constituent two-way interactions, to explicitly assess for any differential effects of age. Intervention group was a between-subjects factor and time was a within-subjects factor. We examined the assumptions of ANCOVA prior to analysis and found no substantial violations. Missing data were handled using listwise deletion. Analyses were run in SPSS (version 31).

Prior to conducting our main analyses, we assessed for demographic differences by treatment conditions and dropout status. Demographic variables included age at baseline, years of education, sex, and race/ethnicity. We also performed preliminary analyses assessing differences in our outcome variables by dropout status. Finally, we tested whether the above listed demographic variables and a series of proposed covariates (BMI, self-reported participation in regular physical activity, average number of servings per day of alcohol) were associated with each cognitive outcome. Preliminary analyses were conducted in SPSS (version 28). We used α = .01 to control for elevated familywise Type I error due to conducting multiple statistical tests. As a result of these analyses, we controlled for age at baseline across all models.

3. Results

Participants (n = 251) were randomized to either the AVO (N = 125) or CON (N = 126) group, of which 241 (AVO group N = 119; CON group N = 122) completed the 6-month intervention. There were no significant differences in demographic characteristics between those who dropped out (n = 10) and those who completed the study. Among those who completed the study, there were no significant differences in demographic characteristics between individuals in the AVO vs. CON group (Table 2). There were no consistent relationships between any of these variables and cognitive outcomes over time, except for age at baseline, which was inversely correlated with all outcomes at each time point, ps < 0.01. Therefore, we controlled for baseline age in all models. We performed a power analysis using GPower, which indicated that we had an 80% chance of detecting a small-to-medium effect size of at least f = .18 given a minimum sample size of 236, two independent variables, and one covariate at alpha = .05 [24].

Table 2.

Sample characteristics by intervention group.

	Intervention group
Demographic characteristics	Avocado group N = 119	Control group N = 122
	M(SD)
Age at Baseline, years	49.11 (13.23)	50.05 (13.50)
Education, years	14.81 (2.34)	14.47 (2.49)
Body Mass Index, kg/m2	33.17 (5.66)	33.67 (5.81)
	N(%)
Age at baseline
25–39 years	37 (31.1)	28 (23.0)
40–59 years	58 (48.7)	66 (54.1)
60–84 years	24 (20.2)	28 (23.0)
Sex
Female	84 (70.6)	84 (68.9)
Male	35 (29.4)	38 (31.1)
Race
Caucasian	69 (58.0)	59 (48.4)
Black/African American	12 (10.1)	15 (12.3)
Hispanic	33 (27.7)	36 (29.5)
Asian	4 (3.4)	9 (7.4)
American Indian/Alaska Native	0 (0.0)	3 (2.5)
Other	1 (0.8)	0 (0.0)

Note. Group differences were non-significant across all variables.

There were no significant differences between the AVO and CON groups in self-reported avocado intake at baseline, t(111) = .186, p = .853. Dietary compliance was excellent in our sample. Every subject in the AVO group reported avocado consumption at any of the 3 unannounced 24-h dietary recalls during the intervention, with >97% of the dietary recalls reporting consumption of one avocado that day. None of the subjects in the CON group reported consuming one avocado on any of the 3 dietary recalls, but in about 12% of the recalls small amounts of avocado consumption were reported. At end of intervention, the AVO group reported a higher intake of avocados, as well as energy, dietary fiber and MUFA, and slightly lower carbohydrate and protein intake at end of intervention compared to the CON group.

3.1. Cognitive outcomes

There was a significant main effect of time for working memory reaction time, such that working memory reaction time was significantly faster for all participants at the end of intervention (M = 0.670, SD = .015) compared to baseline (M = 0.658, SD = 0.144; F[1, 232] = 5.050, p = .026), but there was no group difference between AVO vs. CON (Table 3). There were no significant differences in memory, processing speed, executive function, or simple reaction time by intervention group or time, nor were there any significant group x time interactions across all 5 domains (see Table 4 for estimated marginal group means).

Table 3.

ANCOVA models examining effects of time, intervention group and age on cognitive performance.

	Memory	Processing speed	Executive function	Simple RT	Working memory RT
	F	F	F	F	F
Time	1.067	0.700	2.267	0.880	5.050*
Group	0.018	0.086	0.004	1.281	0.470
Age	41.614**	45.467**	24.672**	13.812**	37.110**
Time * Group	0.707	1.298	0.091	0.014	3.157
Time * Age	2.066	0.000	0.189	0.009	3.330
Group * Age	0.126	0.029	0.149	2.742	0.941
Time * Group * Age	0.744	0.391	0.231	0.191	3.891

Note. RT = Reaction Time.

^*p < .05.

^**p < .001.

Table 4.

Estimated marginal means for cognitive outcomes by intervention group and time, and controlling for age at baseline.

Intervention group	Time	Memory	Processing speed	Executive function	Simple RT	Working memory RT
Avocado	Baseline	0.698	0.809	0.645	0.817	0.668
	6 months	0.690	0.813	0.654	0.778	0.677
Control	Baseline	0.685	0.777	0.630	0.778	0.648
	6 months	0.677	0.796	0.642	0.757	0.663

Note. Means are expressed as proportion of maximum scaling scores, with higher scores representing better performance.

RT = Reaction Time.

Age was a significant covariate across all analyses, ps < 0.001. However, there were no significant two- or three-way interactions with age (time × age, group × age, or time × group × age). Effect sizes for all non-significant findings were very small (all partial η² values ≤ .016), suggesting that our null results were likely not due to Type II error.

4. Discussion

In this HAT ancillary study of adults with central obesity, daily consumption of one avocado for six months did not result in significant changes in cognitive performance. These findings contribute to the growing literature evaluating dietary strategies during midlife that may influence trajectories of cognitive aging. Only two prior trials evaluated avocado intake and cognition. They similarly reported either null findings or modest, domain-specific benefits, particularly within executive function. For instance, in a 12-week randomized controlled trial involving younger adults aged 25–45 years with overweight or obesity, daily consumption of one avocado a day improved performance on a measure of attentional inhibition, a component of frontal-executive function compared to an isocaloric control meal matched for macronutrient composition [20]. However, no significant effects were observed for response inhibition, suggesting that any cognitive benefits associated with avocado intake may be domain-specific rather than global. Individuals with lower BMI often show stronger lipid and cognitive benefits from consumption of unsaturated rich foods like nuts, whereas these effects are often diminished in people with overweight or obesity [25,26]. This pattern suggests that body composition and metabolic phenotype may influence dietary responses, due to obesity-related metabolic alterations or differences in nutrient metabolism [27].

Beyond body composition, age, baseline metabolic status, and cognitive reserve influences responsiveness to dietary interventions. Aging is accompanied by physiological changes including alterations in lipid metabolism, oxidative stress, and neuronal plasticity that can modify the uptake and utilization of bioactive nutrients. In a 6-month RCT of avocado consumption among healthy-weight, cognitively normal older adults aged 50 years and above, no significant group differences were observed over time in attention, visual memory, or executive function scores [19]. Commensurate with that finding, both groups showed gains in spatial working memory and overall memory. Only one measure, problem-solving efficiency and sustained attention, was reported to be improved in the avocado but not control group. In a retrospective study, older adults who reported consuming avocado or guacamole in two 24-h dietary recalls had higher memory and global cognition scores compared to non-consumers [28]. Therefore, findings from past studies suggest that avocado intake may confer selective cognitive benefits depending on age, with more significant benefits observed in older adults. Our study, however, did not find evidence of differing outcomes based on age.

Given the broad age range of participants, we considered whether age might influence responsiveness to avocado consumption, but the pattern of findings did not support differential cognitive effects across young, middle-age and older adults. Evidence from our prior work [29] suggests that nutrient-dense foods may yield greater cognitive benefits in older individuals with elevated cardiometabolic risk. Because our participants were overweight yet otherwise metabolically healthy, this may have limited the cognitive effects observed with avocado consumption. Moreover, even among older adults, some studies suggest that the benefits are more pronounced in the oldest old, aged 70 and above [30], which only comprised 10% of our sample. The limited effects observed here may therefore reflect the earlier stage of cognitive aging in midlife, when detectable benefits of single-food interventions are likely to be modest.

Although participants demonstrated high compliance with their assigned dietary protocols, the quantity of bioactive nutrients provided through avocado consumption may have been insufficient to elicit measurable cognitive effects. The minimum effective dose of avocados or their key bioactive components for improving cognition remains unclear. A systematic review concluded that dietary supplementation with lutein and zeaxanthin enhanced cognitive performance and visual processing speed [31]. A single avocado provides about 360 μg of combined lutein and zeaxanthin [11], a dose 20-fold lower than that used in most supplementation trials (10−12 mg). Therefore, the modest intake of these carotenoids from one avocado per day may not be sufficient to produce the tissue concentrations that have been suggested in limited studies to influence neural function over a six-month period.

The strengths of this study include its large sample size, the use of a comprehensive neurocognitive battery assessing multiple domains of cognition, and the 6-month intervention duration, which exceeds that of most prior avocado trials (12 weeks), although in individuals with obesity, even longer duration may be needed. The inclusion of adults across a broad age range also enabled the exploration of potential age-related interactions and the detection of differential treatment responses across the lifespan. A key limitation, however, is that as an ancillary study, participant eligibility criteria were determined by the parent HAT trial. Consequently, the sample was restricted to adults with central obesity and did not include individuals who had normal body weight, a group that may exhibit a greater neurocognitive response to dietary interventions, as suggested by prior evidence.

In conclusion, consuming one avocado per day for six months, without other prescribed dietary or lifestyle modifications, did not significantly alter cognitive performance in adults with central obesity. Although no cognitive benefits were observed, avocados remain a nutrient-dense food rich in monounsaturated fatty acids, fiber, and carotenoids such as lutein, which have been linked to cardiometabolic and neural health in prior research. Additional work is needed to determine whether avocados, as part of dietary strategies initiated in midlife, contribute to healthy cognitive aging, particularly in normal weight individuals who may have other metabolic vulnerabilities.

Author contributions

The author’s responsibilities were as follows: G.L. designed the research, oversaw data collection, and wrote parts of the manuscript; H.M. analyzed the data and wrote parts of the manuscript; M.A. conducted the research, collected data, and wrote parts of the manuscript; J.S. and S.R., designed the research, oversaw data collection, and provided critical feedback on the manuscript. J.S., D.M.R., P.M.K.E., K.S.P., A.H.L., N.R.M., Z.L., and S.R. designed the multicenter HAT. All authors interpreted the findings and have read and approved the final manuscript.

Declaration of Generative AI and AI-assisted technologies in the writing process

This work was prepared without the use of any generative AI or AI-assisted technologies.

Funding disclosure

This study was funded by the Hass Avocado Board in Mission Viejo, California,

Data availability

The data presented in this study can be made available upon request to the corresponding author pending approval from the HAT Steering Committee.

Declaration of competing interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests:

Grace J. Lee, Joan Sabate, David M. Reboussin, Penny M. Kris-Etherton, Kristina S. Petersen, Alice H. Lichtenstein, Nirupa R. Matthan, Zhaoping Li, Sujatha Rajaram report financial support was provided by Hass Avocado Board Avocado Nutrition Center. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.