Acute effects of ayahuasca on neuropsychological performance: differences in executive function between experienced and occasional users
This study (n=24) found that experienced users showed less detrimental effects on higher cognition following ayahuasca intake than occasional users.
Authors
- Jordi Riba
- José Carlos Bouso
Published
Abstract
Background: Ayahuasca, a South American psychotropic plant tea containing the psychedelic 5-HT2A receptor agonist N,N-dimethyltryptamine, has been shown to increase regional cerebral blood flow in prefrontal brain regions after acute administration to humans. Despite interactions at this level, neuropsychological studies have not found cognitive deficits in abstinent long-term users. Objectives: Here, we wished to investigate the effects of acute ayahuasca intake on neuropsychological performance, specifically on working memory and executive function. Methods: Twenty-four ayahuasca users (11 long-term experienced users and 13 occasional users) were assessed in their habitual setting using the Stroop, Sternberg, and Tower of London tasks prior to and following ayahuasca intake. Results: Errors in the Sternberg task increased, whereas reaction times in the Stroop task decreased and accuracy was maintained for the whole sample following ayahuasca intake. Interestingly, results in the Tower of London showed significantly increased execution and resolution times and number of movements for the occasional but not the experienced users. Additionally, a correlation analysis including all subjects showed that impaired performance in the Tower of London was inversely correlated with lifetime ayahuasca use. Conclusions: Acute ayahuasca administration impaired working memory but decreased stimulus-response interference. Interestingly, detrimental effects on higher cognition were only observed in the less experienced group. Rather than leading to increased impairment, greater prior exposure to ayahuasca was associated with reduced incapacitation. Compensatory or neuromodulatory effects associated with long-term ayahuasca intake could underlie preserved executive function in experienced users.
Research Summary of 'Acute effects of ayahuasca on neuropsychological performance: differences in executive function between experienced and occasional users'
Introduction
Bouso and colleagues situate this study within renewed scientific interest in serotonergic psychedelics as both models for mental disorders and potential therapeutic tools. They note that classic tryptamine psychedelics such as psilocybin and N,N-dimethyltryptamine (DMT) act primarily at 5-HT2A receptors, which are densely expressed in prefrontal cortical pyramidal neurons, and that activation of these receptors modulates frontocortical glutamatergic activity. Neuroimaging studies have reported frontal hyperactivity after acute psychedelic administration, yet earlier neuropsychological work in abstinent long-term ayahuasca users found no cognitive deficits and in some tests even superior performance compared with controls, leaving uncertainty about acute cognitive effects and the role of prior exposure.
Methods
Statistical analysis used general linear models (GLM) with drug intake (pre vs post) as the within-subject factor and group (experienced vs occasional) and sex as between-subject factors; age was included as a covariate because groups differed in age. Pairwise comparisons employed Student's t tests and Pearson correlations examined associations with lifetime ayahuasca use; lifetime sessions were log-transformed to reduce skew. To assess possible practice effects, a control experiment was conducted in which 10 drug‑naive individuals repeated the same battery with a 2‑hour interval in between.
Results
The control experiment with 10 drug‑naive participants who repeated the battery with a 2‑hour interval revealed no statistically significant practice-related improvements on the Sternberg or Tower of London tasks; in the Stroop test a trend toward increased difference in errors at retest was observed. The investigators interpreted these control data as providing no evidence that simple practice alone would explain the pattern of performance changes seen after ayahuasca.
Discussion
Several limitations are acknowledged. Testing took place in field conditions rather than a laboratory, all participants completed the drug‑free assessment first which could permit training effects, and the Tower of London may be susceptible to practice because the same items were reused (post‑drug testing used only three difficult problems to mitigate this). The control experiment did not show learning effects sufficient to account for the main findings, but the authors characterise the results as preliminary. They conclude that acute ayahuasca produces mixed effects on cognition and that the relationship between lifetime exposure and reduced acute impairment warrants further study, ideally combining neuropsychological testing with neuroimaging to assess functional brain mechanisms.
View full paper sections
INTRODUCTION
Drugs inducing transient modified states of consciousness are receiving renewed attention as models of disease and also as potential therapeutic tools. Serotonergic psychedelics containing the tryptamine moiety such as psilocybin (4-hydroxy-N,Ndimethyltryptamine) and N,N-dimethyltryptamine (DMT) have been shown to act as serotonin (5-HT) 2A/1A/2C receptor agonists, with their behavioral effects most likely related to activation of the 5-HT 2A receptor sites. Because of the profound impact of psychedelics on thought processes, their recently uncovered capacity to stimulate intrinsic activity within the prefrontal cortex and the similarities between the subjective effects they induce and endogenous psychoses, psychedelics are being proposed as tools to investigate the role of 5-HT 2A receptors in higher cognitive function). Performance of the most complex aspects of human cognition relies on the adequate functioning of prefrontal and other cortico-subcortical networks involved in cognitive control. Sophisticated neuropsychological skills such as working memory and executive functions, which are responsible for performance monitoring, set-shifting, planning, error detection, and inadequate response inhibition, are subserved by these networks. 5-HT 2A receptors are highly expressed in pyramidal cells of the prefrontal cortex. Most glutamatergic neurons in layers II-V express 5-HT 2A receptors (86-100 %), with a maximum (almost 100 %) in layers. Serotonergic modulation of neural activity in the prefrontal cortexis involved in performance monitoring, working memory, response execution, and executive function, suggesting a potential role for the 5-HT 2A receptors in these processes. Studies in animals have shown that 5-HT 2A receptor activation induces excitatory postsynaptic potentials in the frontomedial cortex, an effect mediated by glutamate release. Using electrophysiological recordings of brain slices,identified a subpopulation of layer V neurons in the prefrontal cortex that are excited and depolarized following psychedelic drug administration. The authors showed that the 5-HT 2A facilitates intrinsic network activity independently from thalamocortical afferents. In line with these data, neuroimaging studies using SPECT and PET have found patterns of frontal hyperactivity in humans after administration of mescaline, and the DMT-containing preparation ayahuasca, described below. A recent study using magnetic resonance imaging has corroborated the involvement of the medial prefrontal/anterior cingulate cortex in the effects of psychedelics. Studies in humans on the acute effects of psychedelics on various cognitive tasks have shown that these compounds can disrupt sustained attention, whereas contradictory results have been obtained regarding their capacity to alter spatial working memory. In a recent investigation, we assessed neuropsychological performance in a large sample of long-term ayahuasca users. This DMT-containing plant decoction is typically obtained by boiling together the stems of the vine Banisteriopsis caapi and the leaves of Psychotria viridis. The resulting tea contains β-carboline alkaloids with monoamine oxidase-inhibiting properties that block the metabolic breakdown of the labile DMT and render it orally active. The acute subjective effects of ayahuasca are characterized by changes in the conscious state, which include perceptual modifications in the visual, auditory, and tactile spheres. Visions with eyes closed are frequently accompanied by intense emotions and increased rate of thinking. These effects start after 45-60 min after administration, reaching their maximum intensity between 90-120 min and return to baseline after 4-6 h. The overall pattern of effects is in line with that of other serotonergic psychedelics, such as mescaline, intravenous DMT). In the long-term user study, we tested a sample of 127 ritual ayahuasca users who were drug-free following a period of abstinence from the drug and compared it with 115 controls. Interestingly, the long-term users scored better than the controls on the Stroop test, the Wisconsin Card Sorting test, and the Letter-Number Sequencing task of the WAIS-III. Thus, despite the fact that the prefrontal cortex is a key region targeted by psychedelics, chronic use of ayahuasca was not associated with specific impairment using these neuropsychological tests. In the present study, we wished to investigate the acute effects of ayahuasca on neuropsychological performance, specifically on working memory and executive function tasks that are subserved by cognitive control networks. As explained below, we administered a battery of neuropsychological tests before and during the acute effects of ayahuasca. Given the lack of detrimental effects observed in our previous study in abstinent long-term users, we also wished to explore the role of prior experience with ayahuasca in the degree of impairment observed. We postulated that experienced users would be more able to correctly perform the neuropsychological tasks than their less experienced counterparts.
MATERIALS AND METHODS
Participants and study procedure Twenty-four individuals (12 female) participated in the study. To assess the role of prior ayahuasca use in current neuropsychological performance after acute ayahuasca intake, participants were classified as either experienced or occasional users (see Results section below). Based on the typical pattern of ayahuasca use of twice a month or more practiced by regular ayahuasca users, we defined an occasional user as a person who had taken ayahuasca between 8 and 60 occasions (once every 3 weeks in the past 3 years). Participants exceeding this value were considered experienced users. All participants had abstained from ayahuasca for at least 15 days before the first of two experimental sessions, i.e., the pre-ayahuasca assessment and the subsequent retest 24 h later in the course of an ayahuasca session (post-ayahuasca assessment). Participants were evaluated in the same setting, where they usually took ayahuasca. Urine samples were collected to exclude subjects testing positive for other psychoactive drugs and/or alcohol. Years of education were recorded for each participant as were their scores on verbal and fluid IQ tests. The verbal IQ test used was a Spanish version of the NARTknown as TAP-"Test de Acentuación de Palabras" (Word Accentuation Test). The fluid IQ test used was a computerized version of the matrix reasoning from the WAIS-III. All participants signed an informed consent prior to participation. The study was approved by the ethics committee at Hospital de la Santa Creu i Sant Pau, Barcelona, Spain. The ayahuasca-taking session started at 12:00 noon, and all subjects took a single oral dose of ayahuasca in a glass containing about 100 ml of the tea. Presence of DMT and other typical ayahuasca alkaloids was later confirmed by thin layer chromatography. Two hours after ayahuasca intake, coinciding with the reported time-window of maximum effects, participants were assessed in a separate and quiet room. Computerized tasks were executed on portable computers under the supervision of one of the authors (JR), while another author (JCB) administered the noncomputerized test (see below). Immediately before testing started, participants were asked to rate the intensity of the ayahuasca-induced subjective effects from 0 to 100. The order of administration of the tests was counterbalanced between participants, and the total time to complete the battery was around 20 min.
THE STERNBERG WORKING MEMORY TASK
A computerized version of this verbal working memory test based onwas used. A series of consonants was shown in succession on a computer screen for 1 s. The length of each series varied between three and eight letters across trials. After each series, a fixation cross appeared on the screen for 1 s followed by the target letter. The target letter was shown on the screen until a response was given. Subjects had to indicate whether the target was present or not in the consonant series by pressing a button. The probability of the target letter having been shown in the series was set at 50 %. After each button press, feedback was given as to the correctness and speed of the response. Two initial training trials were presented followed by 42 test trials (14 trials showing series of 3 elements, 14 trials showing series of 5 elements, and 14 trials showing series of 8 elements). Trial type order was randomized. Whenever a subject gave an incorrect answer, an additional trial of the same type was presented, so that the total number of correctly responded trials was the same for all participants (42 trials). Subjects were given feedback after each of the two training trials (their reaction time in milliseconds) and were encouraged to respond in less than 2,000 ms. Reaction time (RT) for correct responses and performance quality (number of errors) was evaluated. Previous research has shown that RT and errors increase the longer the series. Overall task duration was around 10 min. Neuroanatomically, this task recruits the ventrolateral and the dorsolateral prefrontal cortex (D. A recent neuroimaging study found activation in the inferior frontal and anterior cingulate regions, among others.
THE STROOP COLOR AND WORD TEST
A modified computerized version of the taskwas used. Color names were written either in the same or a different colored ink to that denoted by the name shown on the computer screen. Participants had to press one of three different color keys: red, green, or blue according to the color of the ink the word was written in. Each word was shown for 1 s. A total of 120 words were presented: 60 color-name congruent words and 60 color-name incongruent words. Stimuli were presented randomly, and overall task duration was 5 min. The test measures selective attention, cognitive flexibility, conflict monitoring, and resistance to interference. Performance (number of errors) and reaction time measures in compatible and incompatible conditions as well as the difference between them were computed. Neuroanatomically, this task has been associated to activation in the anterior cingulate cortex and the dorsolateral prefrontal cortex (DLPC).
THE TOWER OF LONDON
This task developed bymeasures different components of executive function such as planning, inhibition, impulsivity, and working memory. Participants were requested to arrange colored beads in pegs using the least number of movements and the least time possible. Beads can be moved according to certain rules only. To reduce the total duration of the task in the postayahuasca assessment, subjects were required to solve only the three most difficult problems (numbers 10, 11 and 13) of the total 13 problems used in the pre-ayahuasca assessment the previous day. Target variables were latency time (i.e., time to the first movement), execution time (i.e., total time needed to solve a given problem), resolution time (i.e., the difference between total time and latency time), number of movements for each problem, and the differential number of movements (i.e., the difference between the number of movements and the stipulated minimum required to solve the problem). Neuroanatomically, the network activated during this task includes the DLPC, the anterior cingulate cortex, and parietal regions. The test took less than 5 min to complete per participant.
STATISTICAL ANALYSIS
Mean pre-and post-ayahuasca scores and standard deviations (SD) were calculated from individual scores for each target variable and are reported as summary statistics. Individual data were subjected to a general linear model analysis (GLM) with drug intake (pre vs. post) as a withinsubjects factor and group (experienced vs. occasional) and sex (male vs. female) as between-subjects factor. Pairwise comparisons were performed using Student's t test (betweenor within-subjects as required), and correlations with lifetime ayahuasca use were calculated using Pearson's correlation coefficient. Given the statistically significant difference in age between experienced and occasional users (see below), age was included as a covariate in the analysis. Differences were considered statistically significant for p values lower than 0.05.
DEMOGRAPHICS
Demographic data are shown in Table. The overall study sample consisted of 24 participants (12 women), with a mean age of 46 years (range 29-62). Experienced users (4 men and 7 women) had taken ayahuasca on an average of 179 occasions, ranging from 70 to 352. Occasional users (8 men and 5 women) had taken ayahuasca on an average of 33 occasions (from 8 to 60). Values differed significantly between groups [t(22)=4.86, p=001]. The two subgroups did not differ in years of education [t(22)=-1.12, p=0.28], and all subjects had intelligence scores in the normal range. No differences were found between the two subgroups in performance in the WAIS matrices test t(22)=-0.34, p=0.74 or in the verbal intelligence (TAP) assessment t(22)=-0.10, p=0.92. Gender distribution did not differ between groups χ 2 (1)=1.5, p>0.1. However, age was found to be significantly greater in the occasional users than in the experienced users t(22)=-3.25, p=0.004. To account for this difference as a possible confound, we used age as a covariate in the statistical analysis of scores on the neuropsychological tests.
SUBJECTIVE EFFECTS AND NEUROPSYCHOLOGICAL ASSESSMENT
All participants reported having experienced intense psychotropic effects at the time of the neuropsychological assessment following ayahuasca administration. Scores for subjective effects were statistically different from zero [t(23)=21.3, p<0.001]. As shown in Table, mean intensity of effects was 72 for the experienced users and 77 for the occasional users. Scores were not significantly different between the two subgroups t(22)=-0.65, p=0.52. Mean data for the neuropsychological tests and results of the statistical comparison between pre-and post-administration values are shown in Table. The statistical analysis of the data from the Sternberg task by means of the GLM showed a main effect of drug (pre vs. postadministration) for total errors [F(1,19)=8.75, p=0.008]. More errors occurred under the effects of ayahuasca for the whole sample. No main effects of group (experienced vs. occasional) or sex were observed. However, we found a trend interaction between drug and group [F(1, 19)=3.87, p=0.063]. To look further at the potential differences between groups, the difference value in total errors (postadministration-preadministration) was calculated and compared between groups by means of an independentsamples t test. This test showed a marginally significant effect [t(22)=-2.07, p=0.051], with a larger mean number of total errors in the occasional users than in the experienced ayahuasca users. Given the known effects of age on memory, we tested for any effects of the covariate "age". A trend interaction was observed between drug and age but did not reach statistical significanceData from one subject was lost for the Tower of London. Thus, the GLM was performed for 23 participants. The analysis did not show any main effects of drug (pre vs. postadministration). However, significant interactions between drug and group (experienced vs. occasional) were found for mean execution timeGiven the differences found between groups in the Tower of London, a correlational analysis was conducted between variables in this test (the difference between post and preadministration values) and lifetime ayahuasca use. The number of lifetime ayahuasca sessions for each participant was transformed to the natural logarithm to correct for the skewed nature of their distribution. The log-transformed values showed statistically significant negative correlations with the difference values (postdrug-predrug) for mean execution time (r=-0.607, p=0.002), mean resolution time (r=-0.588, p=003), mean number of movements (r=-0.482, p=0.020), and mean differential number of movements (r=-482, p=0.020). The corresponding scatter plots are shown in Fig.. To rule out any confounds associated with age, partial correlations were calculated controlling for this variable. Effects were maintained in all four instances: mean execution time (r=-0.618, p=0.002), mean resolution time (r=-0.606, p=003), mean number of movements (r=-0.565, p=0.006), and mean differential number of movements (r=-0.566, p=0.006).
CONTROL EXPERIMENT
A control experiment was conducted to assess the influence of practice in the performance of the administered tasks. The same three tests used in the main study were administered twice with a 2 h interval to a sample of 10 drug-naive individuals (7 women) with a mean (SD) age of 40.1(9.4)years, 12.3(2.5)years of education, a 15.7 (3.1) score on the WAIS matrices, and 25.1 (3.8) on the TAP test. No statistical differences were found between the control group and the two main study samples for any of the demographic variables assessed, except age, which was significantly lower than that of the occasional user group [t(21)=3.1, p=0.005]. Mean data for the neuropsychological tests and results of the statistical comparison (Student's t-test) between pre-and post-administration values for the control sample are shown in Table. No statistically significant differences were observed between the first and the second assessment for any of the variables of the Sternberg or Tower of London tasks. In the Stroop test, difference in errors (incompatible-compatible conditions) showed a trend increase at retest. Thus, no evidence was found of learning effects improving performance at retest in the absence of drug administration.
DISCUSSION
In the present study, we found that acute ayahuasca intake led to a disruption of verbal working memory (Sternberg task), with a trend towards higher impairment in the occasional users, whereas stimulus-response interference was decreased. Speed in the Stroop task was increased, and importantly, subjects did not respond more impulsively and inaccurately. Mean values of incorrectly answered incompatible stimuli were lower after ayahuasca in both participant groups and the percentage of omitted responses showed a statistical trend to decrease. Higher cognition, as measured by the Tower of London, was selectively affected in the occasional ayahuasca users but not in the experienced subgroup. Prior research on the cognitive effects of psychedelics has found disruptions of spatial working memory after psilocybin in some studiesbut not in others. Psilocybinand DMTconsistently impaired performance in sustained attention), visual-spatial attention, and alertness. Psychedelic tryptamines are known to disrupt processes such as binocular rivalry), global motion perception, time perception and temporal control, time interval reproduction, model object completion, and inhibition of return). However, not all studies show detrimental effects. Whereas the mismatch negativity was found to be blunted after DMT, it was not after psilocybin. Also, indirect semantic priming was increased after psilocybin. The present findings show mixed effects of acute ayahuasca on neuropsychological performance. Whereas the Stroop task showed improvements, effects on the Sternberg and on the Tower of London were detrimental, although only for the less experienced users. These results raise the possibility that experienced users may have developed mechanisms to compensate for the acute impairing effects of ayahuasca on executive function. This possibility is supported by the negative correlations found between performance in the Tower of London and lifetime ayahuasca use. Rather than increasing with previous exposure to ayahuasca, the detrimental effects of acute ayahuasca on this task were decreased. Paradoxical results have been reported for other psychoactive drugs, such as tetrahydrocannabinol (THC). Several studies have found only subtle impairments in complex cognitive performance when THC was administered to experienced marihuana users. However, these experienced subjects showed alterations in neurophysiological measures when compared to infrequent users.postulated the implementation of compensatory mechanisms as an explanatory mechanism of maintained cognitive performance in inhibition of return, episodic memory, working memory, and verbal fluency after ketamine. It is worth noting that in most studies reporting acute drug-induced impairment, participants had little prior experience with the administered drug. The observed resistance to impairment or compensatory effect could be due to various mechanisms, both psychological and neurobiological. It cannot be ruled out that tolerance to acute drug effects or mere familiarity with the modified state of awareness induced by ayahuasca could account for the observed results. However, we believe that the tolerance account is unlikely, considering that the intensity of subjective effects was not significantly different between participant subgroups. Familiarity could have played a role in the differential impairment, with experienced users becoming less distracted by the acute effects of ayahuasca while performing the neuropsychological tasks. Nevertheless, we were careful not to recruit drug naive individuals for the study. The occasional users were also familiar with the effects of ayahuasca, having consumed it on average on 33 occasions. On the other hand, recent neuroimaging studies comparing chronic cannabis users with controls show the capacity of long-term psychotropic drug use to induce compensatory functional mechanisms. Cannabis users have been found to exhibit greater prefrontal activations when conducting attention) and inhibition tasks; and greater functional connectivity between frontal and occipitoparietal brain regions in an interference task. Analogous functional mechanisms could potentially be responsible for compensation in long-term psychedelic drug users. An alternative explanation is that chronic ayahuasca intake could lead to neural changes that might help to cope with novel tasks and settings facilitating learning. Hallucinogens stimulate c-fos expression in the medial prefrontal and anterior cingulate corticesand dramatically increase the expression of brainderived neurotrophic factor (BDNF) in the prefrontal cortex. In turn, BDNF influences synaptic efficacy and neural plasticity. As the Tower of London is the test with potentially larger task-related learning effects (especially considering that the same items were repeated), faster learning effects could underlie the beneficial effect observed in the experienced users, indicating better cognitive flexibility or adaptability to new task-settings or problems. We should acknowledge several limitations of the present study. The investigation was not conducted in a laboratory but under field conditions. Furthermore, the order in which participants were tested was not balanced, and they all did the tests in the drug-free condition first. This could potentially have led to learning effects. However, this is unlikely for the Sternberg and Stroop tasks, where the stimuli are randomly presented each time the test is run. Performance in the Tower of London is more susceptible to training effects, but this was probably minimized by our using only three of the most difficult problems. Results should thus be considered preliminary. In any case, even if learning effects were present, these did not prevent us from detecting detrimental effects in working memory and a selective impairing effect in the Tower of London. It is also worth noting that the control experiment we conducted a posteriori did not show any significant learning effects from the first to the second of two consecutive assessments. To conclude, acute administration of the psychedelic DMT-containing beverage ayahuasca led to mixed effects on neuropsychological performance, negatively affecting working memory but not stimulus-response interference. Detrimental effects on higher cognition, as measured by the Tower of London, were observed in the occasional users but not in the experienced users. Negative correlations between performance in this task and lifetime ayahuasca use do not support a priori an association between chronic use and impaired cognitive abilities. This raises the question of whether direct drug-induced neuromodulatory effects or compensatory mechanisms are present in this population. Future investigations should ideally implement neuroimaging techniques to assess brain function during the execution of cognitive tasks in long-term ayahuasca users.
Full Text PDF
Study Details
- Study Typeindividual
- Populationhumans
- Characteristicsobservational
- Journal
- Compound
- Authors