Changes in primitive reflexes in older adults and their relationship to mental health indices: An experimental investigation

The reemergence of primitive reflexes (PRs) in older age is related to cognitive impairment. Currently, there are no means to prevent or slow their reappearance, but research evidence exists for their control in children. Therefore, this experiment investigated whether a 16-week special sensorimotor exercise program could benefit older adults and whether the intervention-induced changes (if any) may be associated with various indices of mental health. Of 115 adults over 60, 95 completed the study (mean age = 76.37 ± SD = 7.04 years, 22 % men). The experimental group ( n = 38) showed an almost threefold decline in PRs compared to controls. In contrast, the control group ( n = 57) exhibited a nearly threefold increase in PRs compared to the intervention group. Cognitive function increased in the experimental but not in the control group. Changes in PRs over the 16-week intervention were positively related to negative mental health indices (hopelessness and perceived stress) and negatively related to well-being. These findings suggest that the here-presented mild sensorimotor exercises could affect older adults ’ reemerging PRs and that changes in PRs are associated with mental health benefits. These results may open new research avenues toward preventing cognitive and psychological decline in older adults.


Introduction
Primitive reflexes (PRs) are automatic responses mediated by the brainstem, which gradually come under cortical control as the nervous system matures, a process regulated by gamma-aminobutyric acid (GABA; Ditmar, 2011;Laliberte et al., 2022;Zafeiriou, 2004).GABA levels are low at birth, peak in early adulthood, and decline after age 50, potentially leading to the reemergence of these reflexes in older adults (Porges et al., 2021).Brain-Derived Neurotrophic Factor (BDNF) and GABA are crucial for neuronal function and balance.Reduced BDNF in aging contributes to GABA dysregulation and neurodegeneration (Kim et al., 2017).Decreased physical activity with age further lowers BDNF, weakening cortical inhibition and potentially triggering PRs (Sleiman et al., 2016).Understanding the pathophysiologic basis of the link between PR and exercise is critical in studying these reflexes in aging populations.
Primitive reflexes are intricate, involuntary movement reactions governed by brainstem structures.They start to develop around the 25th week of gestation and typically fade after the first six months of life (Zafeiriou, 2004).However, they could reemerge with aging due to neurodegenerative diseases or without diagnosed brain dysfunction (Boxtel et al., 2006).Their disappearance in early childhood is associated with the maturation of neural networks, particularly in the frontal brain regions, leading to improved neural and motor skills (Isakov et al., 1984;Pecuch et al., 2021).In older adults, PRs might suggest diminished cortical regulation by lower brain centers, potentially resulting from neural network damage due to cerebral degeneration, such as white matter hyperintensities or ischemic injuries (Boxtel et al., 2006).This reemergence of PRs is sometimes called 'inverse development,' which is especially common in patients with senile dementia or Alzheimer's disease (Altunkalem Seydi et al., 2024).While it is uncertain whether PRs reappear as a natural part of aging, the 66-82 age group exhibited the highest PR rate in the Maastricht Aging Study (Boxtel et al., 2006).
The prefrontal neural networks, critical in complex behaviors such as planning, working memory, and task execution, are particularly vulnerable to aging.Cognitive decline and the presence of PRs may share underlying neurodegenerative mechanisms (Damasceno et al., 2005).Research with older adults and dementia patients has linked the presence of PRs to the severity of dementia, though the results are inconsistent.However, a meta-analysis revealed that the occurrence of PRs in older adults with dementia was 13.94 to 16.38 times higher than in healthy controls (Altunkalem Seydi et al., 2024).Although studies explicitly examining the presence of PRs in healthy adults are limited, recent research found that 33.1 % of adults aged 45 to 91 exhibited PRs, with prevalence increasing with age.These individuals performed worse on cognitive tasks than those without PRs (Camarda et al., 2019).Therefore, a link between PRs and cognitive functioning is substantiated in the literature.
While the relationships between PRs and mental functions are correlational, a decrease in PRs-when present-could be expected to weaken the strength of such relationships.Thus, whether PRs' inverse development could be reversed or halted may be a justified question.A special sensorimotor program to eliminate or reduce PRs has already been successfully applied to children (Stephens-Sarlós et al., 2024).Whether such training is also effective in older adults is to be determined.Indeed, this research path is yet exploratory because the neuroplasticity in children and older adults differs.This fact is substantiated by Park and Bischof (2013), who summarize their literature review by stating, "Evidence suggests that older adults show less neuroplasticity than younger subjects, and we argue that interventions that successfully delay agerelated cognitive decline will yield greater benefits than short-term facilitation of cognition."(p.117).Despite differences in neuroplasticity, these authors trust that interventions that counteract age-related cognitive impairment exist.
Primitive reflexes were not studied in association with mental health indices such as perceived stress, hopelessness, well-being, happiness, or satisfaction with life but only with the dysfunctional aspects of mental functions.Zafeiriou (2004) summarized adult PRs as frequent indices of frontal lobe damage or upper motor neuron dysfunctionality and markers of cortical disinhibition.He also identified the clinical morbidities in which PRs are used in neurological examinations, such as senile dementia, Alzheimer's disease, schizophrenia, multiple sclerosis, Parkinson's disease, frontal lobe lesions, and hydrocephalus.But the opposite may also prevail.Indeed, a PR reduction may be associated with improved mental health indices.For example, changes in PR frequency can be expected to relate positively to negative psychological indices (i.e., stress) and inversely associated with positive measures (i.e., happiness).
This experimental study examined the effects of a 16-week sensorimotor exercise program (Appendix A) developed to control or reduce the reemergence of PRs in older adults.Based on recent research evidence from studies with children (Infante-Cañete et al., 2023;Stephens-Sarlós et al., 2024), despite lower neuroplasticity in aging adults, we hypothesized that a specifically designed exercise program could help reverse or halt the inverse development of PRs and associated cognitive issues.We also hypothesized that changes in PRs (if any) would be inversely related to positive mental health indices, such as cognitive function, well-being, happiness, and satisfaction with life, and directly associated with adverse mental health measures, such as perceived stress and hopelessness in older adults.

Sample size calculation
Using the G* Power software (Faul et al., 2007), we calculated the needed sample size for a 2 (time) by 2 (group) repeated measures multivariate analyses of variance (MANOVA).The G* Power test selection was a) F tests, b) MANOVA, c) repeated measures with between factors, and d) the following input parameters: medium effect size (Cohen's f = 0.25), alpha (α) = 0.05, power (1-β) = 0.80, and correlation between repeated measures (r = 0.50).This calculation yielded a minimal sample size of n = 98.

Participants
Participants were recruited from nine care homes for older adults after obtaining approval from the management of these establishments.Eligibility criteria included being 60 years or older and being able to stand and walk.Exclusion criteria were based on physical conditions such as occasional dizziness, cardiorespiratory disease, loss of balance, and untreated hypertension, as well as psychiatric conditions, including previously diagnosed mental and behavioral disorders.Medical clearance regarding these criteria was obtained for all participants who signed an informed consent form and a General Data Protection Regulation (GDPR) data handling form before participating in the study.In total, 115 older adults agreed to participate, but only 95 (22 % men) completed the study (M age = 76.37 years; SD = 7.04, range 62 to 94).The study was conducted during the spring and summer of 2024 with approval from the Research Ethics Board (Permission No. DHK-2024/ 00039/2) at Széchenyi István University.The study adhered strictly to the principles of the Helsinki Declaration (World Medical Association, 2013) for research involving human participants.

Cognitive function
Before and after the intervention, all participants were screened with the Mini-Mental State Examination (MMSE; Folstein et al., 2001), a widely used 30-point tool for measuring cognitive function.The MMSE covers various cognitive areas, including orientation, word recall, arithmetic, language, comprehension, and basic motor skills.An example task is following a three-step command: "Place the index finger of the right hand on your nose and then on your left ear" (Item 3, language).Despite criticism regarding its diagnostic accuracy, the MMSE is supported for assessing cognitive function in healthy older adults (Truong et al., 2023).We selected this instrument for its comprehensive measurement of cognitive functions and brief administration time (Kurlowicz and Wallace, 1999).The same researcher, assisted by observing a doctoral candidate, both being blind to grouping, conducted all tests.

Well-being (WHO-5)
The WHO-5 (World Health Organization, 1998;Topp et al., 2015) is a 5-item well-being assessment tool that measures an individual's subjective well-being over the past two weeks.Each item is rated on a 6point Likert scale, from 0 (not present) to 5 (constantly present), addressing positive mood, vitality, and general interests.An example item (No. 1) is "I have felt cheerful and in good spirits."The total score ranges from 0 to 25; a higher score indicates better well-being.Its internal consistency (α) across 15 European nations was 0.86, ranging from 0.70 to 0.90 (Cosma et al., 2022).Notable is that the WHO-5 collects retrospective data subject to memory bias, which is a delimitation of the studies using it.

Subjective happiness
The Subjective Happiness Scale (SHS; Lyubomirsky and Lepper, 1999;Szabó, 2019) is a brief, 4-item questionnaire that assesses an individual's overall happiness.Each item should be rated on a 7-point Likert scale, with responses ranging from 1 (less happy) to 7 (much happier).Two items ask respondents to rate their happiness relative to their peers, and two provide descriptions of happy and unhappy individuals, asking respondents to rate how well each description fits them.Most researchers adopt the SHS for its simplicity and good psychometric properties, with higher scores indicating greater subjective happiness.Its internal reliability ranges from Cronbach's alpha (α) 0.65 to 0.94 in over 30 studies from >15 nations (Szabó, 2019).

Perceived stress (PSS-4)
The 4-item Perceived Stress Scale (PSS-4; Cohen et al., 1983) is a brief measure developed to assess perceived stress over the past month.
It includes four questions rated on a 5-point Likert scale, ranging from 0 (never) to 4 (very often).The items address feelings of stress and how unpredictable, uncontrollable, or overloaded one feels.Scores range from 0 to 16, with higher scores indicating greater perceived stress.The PSS-4 has acceptable internal reliability (0.74) based on a large sample from three Western European nations (Vallejo et al., 2018).Like the WHO-5, the PSS-4 collects retrospective data.However, in contrast to WHO-5, the PSS-4 collects data over the past month rather than two weeks, increasing the risk of memory bias while also being affected by momentary psychological states.It was demonstrated that several general/dispositional (not necessarily retrospective) psychological measures are influenced by momentary feeling states when answering (Szabo and Ábel, 2021).

Beck hopelessness scale (BHS-4)
The four-item Beck Hopelessness Scale (BHS-4; Aish and Wasserman, 2001) assesses feelings of hopelessness and pessimism.Each item is rated on a dichotomous scale (true or false), with statements reflecting negative attitudes about the future, self-worth, and the likelihood of positive changes.The total score ranges from 0 to 4, with higher scores indicating greater hopelessness.The BHS-4 is valued for its brevity and effectiveness in screening for hopelessness, a key indicator of suicidal risk.In a large population-wide survey, its internal reliability was Cronbach's α = 0.85 (Perczel Forintos et al., 2013).

Satisfaction with life (SWL)
The Satisfaction with Life Scale (SWLS; Diener et al., 1985;Martos et al., 2014) is a 5-item self-report questionnaire to measure global cognitive judgments of one's life satisfaction.Each item is rated on a 7point Likert scale, ranging from 1 (strongly disagree) to 7 (strongly agree).Example items include statements like "In most ways, my life is close to my ideal."The total score ranges from 5 to 35, with higher scores indicating greater life satisfaction.Its internal consistency in a large sample of adults was (Cronbach's α) 0.86 (Martos et al., 2014).

Assessment of primitive reflexes
We tested primitive reflexes (PRs) before and after the intervention following the methods described by Sanders and Gillig (2011) and Fiorentino (2014).For example, to elicit the grasp reflex, the experimenter applied firm pressure to the participant's central palm, moving to hold on; otherwise, some participants may have assumed that they were expected to do so.The experimenter avoided contact with the back of the hand, as this may have counteracted the reflex.Another example is the Babinski reflex, which was elicited by drawing a somewhat dull object (e.g., a tongue blade) down the lateral sole of the foot, starting just anterior to the heel and proceeding to the lateral ball of the foot, then continuing medially to finish just posterior to the medial ball of the foot.This L-shaped motion required 1 to 2 s.The experimenter noted an extensor response at the base of the first toe.
Before summing them up, 13 left and right body-side reflexes were tested by the same experimenter in six repetitions each, including the symmetrical tonic neck reflex (STNR), asymmetrical tonic neck reflex (ATNR), tonic labyrinthine reflex (TLR), Galant reflex, Moro reflex, palmar grasp reflex, sucking reflex, Babkin response, Babinski reflex, core tendon guard reflex (CTGR), glabellar reflex, Schilder test, and vestibular-ocular reflex.The presence (1) or absence (0) of primitive reflexes (PRs) was scored based on observation in at least four out of six trials, which an observing researcher endorsed.

The 16-week intervention
Before the intervention, PRs were evaluated by one researcher in a consulting room, and then another researcher completed the MMSE evaluation for all participants in another room.These researchers did not know about the grouping, and each of them worked with a doctoral student, seeking agreement with them before recording each PR and MMSE test result.Subsequently, participants were randomized into intervention (odd number, n = 57) and control (even number, n = 58) groups.However, 19 individuals who were allocated to the intervention group and one randomized in the control group withdrew consent after group allocations primarily for inability to commit to exercise sessions or willingness to participate in the other than the allocated group.Therefore, 38 older adults comprised the intervention group and 57 in the control group, all of whom completed the study.The researchers informed control participants that they would undergo testing again after 16 weeks.They would need to complete five brief questionnaires, PR tests and the MMSE at that time.Individuals in the intervention group started the exercise session the day after group allocation.
The special sensorimotor exercises targeted the attenuations of 10 PRs 1) the ATNR reflex, 2) the STNR reflex, 3) the Moro reflex, 4) the Galant reflex, 5) the Palmar grasp reflex, 6) the sucking reflex, 7) the CTGR reflex, 8) TLR reflex, 9) the Glabellar reflex, and 10) the Babinski reflex (Stephens-Sarlós, 2023).The Appendix A presents these exercises, which took place over 16 weeks, twice weekly, and lasted slightly over 30 min each time.They were conducted by a qualified and experienced physiotherapist on the same day of the week, in the morning hours, in the participants' care homes, and the same venue designated for physical rehabilitation.Video assistance and the trainer ensured the correct execution of the movements.

Data analysis
First, we checked for normal data distribution by assessing the skewness and kurtosis of the measures.Besides cognitive function after the intervention, skewness values ranged from − 2.00 to +2.00, and kurtosis values ranged from − 4.00 to +4.00.According to Hair et al. (2010) and Byrne (2010), data are normally distributed if skewness is between − 2.00 and + 2.00 and kurtosis is between − 7.00 and + 7.00.We then calculated a repeated measures analysis of covariance (ANCOVA), using gender and age as covariates, to examine changes in PRs in the two groups over the 16-week intervention period.We also assessed the direction of changes (1 = decrease, 2 = no change, and 3 = increase) in both groups and subjected these data to a Chi-square (χ 2 ) test.Next, given that the normality assumption was not met in cognitive function data after the intervention, we tested the differences between groups before and after the intervention with Mann-Whitney U tests and within groups changes with Wilcoxon signed-rank using the Monte Carlo Exact Tests in both instances.Finally, Pearson's correlations were calculated between the pre-to post-intervention change scores (delta [Δ] scores) and five psychological measures (i.e., well-being.Happiness, perceived stress, hopelessness, and life satisfaction) to test whether such changes relate to mental health indices.

Results
The repeated measures ANCOVA yielded a statistically significant group by time interaction (Pillai's Trace = 0.150, F (1,91) = 16.04,p < .001,effect size [partial Eta squared] = 0.150).The interaction is illustrated in Fig. 1.The time main effect and the effect of the covariates, such as time-by-gender interaction and the time-by-age interaction, were statistically not significant.
Changes in PRs were categorized as decrease (1), no change (2), and increase (3) and subjected to a Chi-square (χ 2 ) test, which yielded a statistically significant result (χ 2 [2] = 17.36, p < .001),indicating that while a decrease in PRs was most common in the exercise intervention group, the opposite (i.e., an increase) was most frequent in the control group.These results are shown in Table 1.
A two-tailed Wilcoxon signed-ranked with Monte Carlo simulation to obtain an exact test, based on 10,000 sampled tables with a starting seed of 1,039,640,005, indicated that MMSE scores were higher in the intervention group after 16 weeks compared to the baseline (z = − 2.44, Monte Carlo exact p = .015,95 % confidence interval, lower = 0.012, upper = 0.019).In contrast, no statistically significant change in MMSE results emerged in the control group (z = − 1.35, Monte Carlo exact p = .179,95 % confidence interval, lower = 0.169, upper = 0.188).Therefore, cognitive function improved only in the sensorimotor exercise group.
Finally, as hypothesized, Pearson's correlation coefficients (r) were statistically significant and negative between changes (difference or delta scores) in the number of PRs and well-being.They also approached but did not reach statistical significance in happiness and life satisfaction.Two statistically significant positive correlations emerged between the change in the number of PRs and perceived stress and between the change in the number of PRs and hopelessness.These results are presented in Table 2.

Discussion
We aimed to test whether a set of sensorimotor exercises practiced twice weekly could affect or decrease the number of PRs in adults aged 60 and over.Based on the significant group-by-time interaction of the repeated measures ANCOVA, we gathered support for our hypothesis since PRs decreased in the experimental group.In contrast, the changes in PR frequency went in the opposite direction in the control group (please refer to Fig. 1).While these results lend support for the influenceability of PRs in older age and appear promising in controlling the negative connection between PRs and neuro-cognitive functions (Altunkalem Seydi et al., 2024;Zafeiriou, 2004), PR changes seem to be complex.Indeed, while about 53 % of participants in the experimental group responded to the sensorimotor intervention, about 26 % showed no change, and unexpectedly, 21 % showed increased PRs (Table 1) despite the intervention.These findings can be attributed to factors other than gender or age because we controlled them in the principal analysis.Nevertheless, undiagnosed health issues or existing psychobiological or physiological degenerations may influence the connection between sensorimotor exercises and their effectiveness in impacting PRs in older adults.
Despite these dynamics in PR changes in the experimental group, the participants in this group achieved better results than controls, whose overall increase in PRs was over 61 %, nearly three times higher than the increased frequency in the experimental group (please refer to Table 1).However, 17.5 % of control participants also experienced a decrease in the number of PRs without participating in the exercise intervention.These findings could result from several factors, including mental and physical states at baseline or temporary minor illness or stress, but the actual reason remains to be determined.

Table 1
The direction of change in the number of primitive reflexes (PRs) in the intervention group (n = 38) and control group (n = 57).In this study, we only expected that our intervention would slow down or decrease the PRs in the experimental group due to the sensorimotor exercise program.We had no hypothesis about changes in the control group we observed in nomothetic (Fig. 1) and idiographic (Table 1) data.Observing an increase in at least one PR in 61 % of control participants was unexpected.Noteworthy is that this figure includes the increase in even 1/10 PR, so they cannot be compared to an overall increase of about 75 % between ages 65-85 (Boxtel et al., 2006).As there are no longitudinal studies of similar duration in this area, we can only speculate that some activities, other than our intervention, might affect some PRs in either direction in shorter periods, like 16 weeks.Therefore, the findings of this work encourage future research to evaluate individual PR profiles beyond comparing mean scores and investigate the circumstances that lead to changes.Since there is a strong connection between PRs and GABA, which declines with age (Porges et al., 2021), measuring GABA with magnetic resonance spectrography (Zuppichini et al., 2024) could aid in a better understanding of changes in PRs over time with and without exercise.
Consequently, individual responses varied despite the relatively straightforward group differences in PR dynamic over the 16 study weeks.Not all people in the experimental group exhibited an improved PR profile.Conversely, not all people in the control group showed increased PRs.Nevertheless, confirming the parametric test, the Chisquare test also supported the difference between the two groups in that more than half of the participants in the experimental group improved their PR profile.In comparison, more than half of the control group worsened their PR profile.The approximately 20 % change in the unexpected direction of PRs in both groups calls for further research identifying the factors that affect the connection between sensorimotor exercises and PR dynamics in older adults.Therefore, our results suggest that the intervention may help reduce PRs in older adults, but it does not establish causality.
In parallel with the promising changes observed in PRs, cognitive functions improved in the experimental but not in the control group.These findings show that the mild sensorimotor exercises developed by these authors (Appendix A) could positively affect both the number of PRs and cognitive functions, thereby strengthening the relationship between PRs and cognitive functioning in older adults (Altunkalem Seydi et al., 2024;Boxtel et al., 2006).Still, it remains unclear whether the favorable changes in cognitive functions occurred because of the intervention or because of an overall decrease in the number of PRs in the experimental group.We speculate that the former explanation may be more feasible since it is doubtful that decreased PRs immediately and simultaneously affect cognitive function.To better understand this complex relationship, neurophysiological studies are necessary, as it is known that there are numerous physiological neurobiological differences between adults with and without PRs (Camarda et al., 2019).The mechanisms involved require substantial systematic research to prevent or delay neurodegeneration associated with the presence of PRs in older adults.
Another novel contribution of this study is that changes in PRs are associated with an overall more positive psychological profile in older adults.For example, the inverse relationship between perceived wellbeing and the number of PRs suggests that a lower number of PRs is associated with higher well-being and the opposite.There was a tendency (that would have been statistically significant with one-tailed tests) of a similar relationship between the number of PRs and happiness and life satisfaction.However, the results did not reach the conservative level of statistical significance with the two-tailed test employed because of the exploratory nature of these connections.It is possible that with a larger sample size, these relationships would be statistically significant, but it should be noted that the shared variances (based on R 2 ) ranged only between 11 %-13 %.Therefore, while changes in PRs are related to psychological indices of mental health, the shared variance appears to be relatively modest but possibly crucial from a practical perspective.
Simultaneously, a direct relationship exists between the number of PRs and the negative indices of mental health, such as perceived stress and hopelessness, indicating that a decreased PR number is favorable.In contrast, an increase in PRs accompanies the worsening of these indices.Because these results are novel, they are incomparable to earlier research.However, the correlations between PRs and psychological measures of well-being, perceived stress, and hopelessness were statistically significant, with a medium effect size (r > 0.30).Theoretically, this value is worth consideration based on an earlier meta-analytic review (Windle et al., 2010) that reported an overall effect size of 0.27 (CI = 0.14 to 0.40) for various exercise interventions on mental well-being (CI = 0.14 to 0.40) and concluded that exercise could improve mental well-being in later life.Nevertheless, since our study employed a newly developed PR-targeting exercise program and because the shared variances were relatively low, the practical value and implications of the results call for further experimental scrutiny.

Limitations
This study has limitations that should be considered.First, despite calculating the prior sample size, the number of participants is relatively small, and the older adults recruited from the eight care homes may not represent the aging population in general.Second, despite prior medical clearance, some participants could have had undiagnosed degenerative disorders or neurologic dysfunctions that might have affected the results.Third, the number of dropouts immediately after randomized group allocation suggests a preference for the intervention, which could be associated with placebo or Hawthorne effects.Fourth, pre-post intervention results might have been influenced by factors other than exercise, but the experimenter has no control over life events over the experimental period.Still, the physical and social environments of the current sample were relatively homogenous because they all lived in residences for older people.Future studies using the sensorimotor exercise intervention presented here should also involve a placebo control group.Fifth, the number of men was underrepresented in this study.Therefore, possible biological sex differences could have remained undisclosed.Sixth, one experimenter assessed the PRs and another cognitive function, assisted by a doctoral researcher to ensure consistency.However, two independent assessments of both measures would yield more reliability despite more laborious data collection and the risk of overtesting, resulting in unwillingness to participate.Finally, we did not control the activities of the control groups, which might have involved actions or routines that could also affect PRs positively and possibly explain the 17.5 % drop in the number of PRs in the control group.

Conclusions
Building on research with children, we designed a gentle sensorimotor exercise program for older adults (Appendix A) and administered it twice weekly over 16 weeks to individuals aged 60 and over.After the intervention, we observed a 52.6 % decrease in PRs and improved cognitive function in the experimental group.In contrast, 61.4 % of the control group exhibited increased PRs with no cognitive improvement over the 16 weeks.The results also reveal, for the first time in the literature, an inverse relationship between changes (delta scores) in PRs and mental health indicators, such as well-being, alongside a direct connection between PRs and adverse mental health outcomes like perceived stress and hopelessness.These findings call for further systematic longitudinal research as they have applied value in preventing or delaying cognitive decline and preserving mental health in older adults.

Fig. 1 .
Fig. 1.Changes in the number of primitive reflexes (PRs) over 16 weeks in two groups.

Table 2 Pearson
's correlations (r) between five psychological measures and change in the number of primitive reflexes (PRs) from baseline to the end of the 16-week training period, the levels of statistical significance (two-tailed test), and 95 % confidence intervals.* Note: Statistically significant.Psychological tests were given the day after testing primitive reflexes (PRs).