Journal Logo


Psychobiological Responses to Preferred and Prescribed Intensity Exercise in Major Depressive Disorder


Author Information
Medicine & Science in Sports & Exercise: November 2016 - Volume 48 - Issue 11 - p 2207-2215
doi: 10.1249/MSS.0000000000001022


People with clinical mental health disorders generally see greater mood benefits from exercise than healthy individuals (28), and exercise is a recommended treatment for depression (37). Rethorst and Trivedi’s (33) evidence-based recommendations demonstrate that a wide variety of exercise programs can be beneficial for depression resulting in recommendations similar to those for the public (i.e., 150 min of moderate- to vigorous-intensity physical activity each week [20]). However, the authors highlight a lack of studies designed to assess the relative effectiveness of different programs. The limited data examining the dose–response relationship of exercise training on depression (14,42) suggest that accumulating more exercise each week generally leads to greater reductions in depression, but this is not consistent across all populations. Therefore, it is possible that alternatives to traditional exercise prescriptions could outperform standard exercise prescriptions for improving depressed mood.

In his seminal work defining and characterizing RPE as a measure of exercise intensity, Gunnar Borg (4) also described the idea of “preferred” exertion, or the level at which someone would perceive their effort to be “just about right or comfortable”—stressing the importance of individual choice. Expectations about the effectiveness of a certain intervention or behavior can affect the resulting perceived efficacy of that intervention (e.g., the placebo effect), particularly when primary outcomes are psychological (43). Therefore, using exercise prescriptions that instruct participants to exercise at an intensity or in a manner that they would not choose might limit the affective benefit of exercise. Failure to account for an individual’s preference could result in a blunted effect of exercise on mood, potentially decreasing the likelihood of future exercise participation (44). However, there is limited data concerning the influence of exercise preference on the affective responses to exercise in depressed individuals.

Callaghan et al. (8) provided the only report the authors are aware of evaluating the effects of exercise at a preferred intensity in an adult population with major depressive disorder (MDD). Although acute responses were not measured, they found that allowing participants to self-select their exercise intensity for a 12-session exercise training program yielded lower posttraining depression scores and greater attendance than participants who exercised at a prescribed intensity. It is plausible that allowing patients to choose their exercise intensity resulted in more pronounced affective responses to each individual bout of exercise, and that this acute response led to better chronic adaptations (i.e., depression scores postexercise training). To our knowledge, the relative affective benefits of preferred exercise compared with prescribed exercise on acute mood responses have not been studied in MDD.

One potential mediator of the exercise–affect relationship, particularly as it relates to the clinical condition of depression, is brain-derived neurotrophic factor (BDNF). The serum content of this protein is increased by exercise (10), and resting levels are typically lower in MDD (6). Moreover, increases in BDNF can be associated with symptom improvement in MDD (36), and resting levels are responsive to some antidepressant treatments (24). Furthermore, it is possible that increases in BDNF levels may be one potential way that exercise improves brain health (30).

Research to date has provided insufficient guidance on how best to prescribe exercise for acute mood improvements in depression. The results of Callaghan et al. (8) suggest, but do not test, that preferred exertion exercise would lead to greater acute responses than prescribed exercise. Currently, it is unknown whether acute psychobiological responses are greater after prescribed intensity exercise or exercise where the intensity is self-selected by the patient. Our purpose was to determine the relative effectiveness of preferred versus prescribed intensity exercise by systematically comparing the affective and biological responses to three set exercise intensities (light, moderate, and hard) and preferred intensity exercise in MDD. On the basis of the work of Callaghan et al. (8) described earlier, we hypothesized that the preferred session would yield superior psychobiological responses to similar prescribed sessions.



Twenty-four women with MDD participated in the present study. Participants were recruited via newspaper advertisements, flyers, and a mass e-mail to the campus community that asked for women with depression to participate in an exercise study. Inclusion criteria were as follows: 1) female, 2) ages 20–60 yr, 3) self-reported physician diagnosis of MDD confirmed via Mini International Neuropsychiatric Interview 6.0 on the first day of testing, 4) safe to exercise based on results from the Physical Activity Readiness Questionnaire (41), and 5) either not on a psychiatric treatment regimen (counseling or medication) or on a stable regimen for the 8 wks preceding their initial visit. We included participants who were currently on stable doses of antidepressant medications as long as they still met the criteria for MDD and agreed to not change their current medication regimen during the study. Because women are diagnosed with depression at much higher rates than men (5), we chose to test only women for this initial investigation. Participants were excluded if they were pregnant, current smokers, diagnosed with concomitant psychological disorders other than generalized anxiety disorder, currently taking opioid or analgesic medications, or currently abusing alcohol or other drugs. Initial eligibility was determined using a phone screen. Participants were compensated a total of $100 for completion of all study visits.


All procedures were approved by the Institutional Review Board at the University of Wisconsin–Madison. These data included in this study are part of a larger protocol designed to examine the acute effects of exercise on depressed mood in MDD and the potential moderators and mechanisms underlying this effect. Detailed results of the dose–response relationship of acute exercise intensity on depressed mood can be found in the study of Meyer et al. (26). The procedures described herein are those that were specific for the preferred exercise intensity aims of the study. Participants reported to the laboratory for four visits, each separated by at least 1 wk, and all participants completed all four testing sessions (light, moderate, hard, and preferred). Before each visit, participants were asked to refrain from performing structured exercise and from drinking alcohol for 24 h, from ingesting caffeine for 4 h, and from using tobacco for 2 h.

Upon arriving for the initial day of testing, participants provided informed consent and were assessed for current MDD and other psychiatric comorbidities via the Mini International Neuropsychiatric Interview (version 6.0.0), a mental health interview designed to diagnose mood disorders based on the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (38). Next, baseline questionnaires (detailed in the following sections) measuring mood and symptoms of depression were completed, and participants self-reported height and weight. Participants then underwent one of the four exercise sessions in a randomized and counterbalanced order.

Each session included 30 min of exercise on a cycle ergometer at one of three prescribed intensities or an intensity of the patient’s choosing. Before exercise and within 10 min after the conclusion of exercise, blood samples were drawn to measure serum BDNF concentration. Self-reported mood was also recorded at 10 and 30 min postexercise to minimize the chance of missing transient or delayed mood responses. After the last questionnaires on the final session, there was an open-ended debriefing session where participants were asked to discuss how they felt during and after the preferred as compared with the prescribed exercise sessions. They also had the opportunity to ask questions about the study and were provided with information regarding how each exercise session influenced their mood.

Exercise sessions

Each exercise session was performed on a stationary bicycle (Lode Corival, Lode BV, Groningen, The Netherlands). Cycle ergometry was chosen as it is safe and a relatively easy mode for participants to produce multiple different levels of intensity. For the prescribed exercise sessions, participants were instructed to cycle at a work rate perceived as light, moderate, or hard based on perceived exertion ratings of “11,” “13,” or “15,” respectively (see Perceived Exertion section). Pedaling rate was maintained between 60 and 70 rpm, and pedal resistance (wattage) was adjusted by the participant to maintain the specified intensity for the session. Exercise began with a 5-min warm-up, during which participants increased their exertion to achieve the session’s intensity by the end of the warm-up. Participants then maintained that intensity throughout the 20-min session and ended with a 5-min cooldown. Watt adjustments were made by the participant to maintain the session’s intensity criterion and were measured with intensity maintenance confirmed by an investigator who was present throughout the session. This protocol is based on RPE production, and similar protocols have been used extensively to allow for the self-regulation of the intensity of exercise by the participant rather than an experimenter-controlled workload (16,17).

For the preferred intensity condition, participants were instructed to choose any intensity of exercise at which they would prefer to exercise. The rest of the exercise instructions were those given in the prescribed sessions with the primary difference being the lack of an intensity designation (see Document, Supplemental Digital Content 1, Prescribed Intensity Exercise Instructions, Although prescribed sessions were performed at a constant subjective intensity, the preferred session did not have this requirement, and participants were free to change their intensity throughout the session.

Exercise equipment and physiological measurement

Before the start of exercise, the seat height and handlebar placement were adjusted for the participants’ comfort. During exercise, measurements of ventilation (V˙E), oxygen consumption (V˙O2), and carbon dioxide production (V˙CO2) were obtained continuously and analyzed with 15-s averaging using a metabolic cart (TrueOne, ParvoMedics, Sandy, Utah) and a two-way non-rebreathing valve (Hans-Rudolph, Kansas City, MO). RER was calculated as the ratio of V˙CO2 to V˙O2. The flowmeter was calibrated before each exercise bout by making multiple comparisons to a 3-L piston syringe. O2 and CO2 measurements were calibrated by the presentation of known concentrations of gas. HR and work rate (in watts) were also assessed and recorded throughout the exercise test and during recovery.

Perceived exertion

Borg’s RPE [4] scale was used for both exercise prescription and assessment of exertion during exercise. Using perceived exertion (RPE) to standardize exercise intensity allows for simple participant-initiated intensity adjustments during exercise (21) is similar to HR for predicting aerobic capacity (18) and is an appropriate comparison for preferred exertion research (13). Perceived exertion was measured at 5-min intervals and workload adjusted by the participant as necessary to maintain the prescribed intensity. In addition, the RPE scale remained in front of the participant throughout the exercise to allow for the easy checking of current perceived exertion and adjustments as needed throughout the 20-min steady-state exercise.

During the preferred session, the participants were allowed to adjust the resistance (W) on the bike continuously in a similar fashion as the prescribed sessions. However, because the preferred session did not have an RPE criterion to maintain, they could adjust their intensity as often or as little as they desired and, in contrast to the prescribed sessions, were not required to maintain any particular intensity during the 20-min session (see Document, Supplemental Digital Content 1, Prescribed Intensity Exercise Instructions, This allowance of variability was intentionally chosen to most closely mimic what a person might choose during an exercise session on their own. This allowed for a more accurate comparison of the effect of traditional steady-state exercise prescription to a self-paced exercise session at a preferred intensity.


The Beck Depression Inventory (BDI) II was used to assess depressive symptoms during the past 2 wk. The BDI is a validated measure of assessing depression symptoms in both psychiatric patients and healthy populations (1). To assess current mood state, we used the Profile of Mood States (POMS)—a frequently used and validated scale to assess mood states within a wide range of populations (25,32). The International Physical Activity Questionnaire —a validated 7-d recall measure of physical activity behaviors (12), was used to measure self-reported physical activity behavior in the last week. To measure current anxiety, participants filled out the State-Trait Anxiety Inventory (state version only)—a validated measure to assess state levels of anxiety (23,39). The BDI, the International Physical Activity Questionnaire, and the State-Trait Anxiety Inventory were filled out preexercise at each session, whereas the POMS was filled out three times per session (preexercise, 10 min postexercise, and 30 min postexercise). The primary dependent variable for the acute effect of exercise on mood was the depression subscale from the POMS.

Blood draw and processing

Whole blood was drawn before exercise and within 10 min of the end of each exercise session. Following standard procedures, blood was allowed to clot for 15 min and then centrifuged for 10 min at 5500 rpm and 4°C. Blood serum was then extracted, and samples were stored at −80°C until BDNF analysis. BDNF analysis was performed via enzyme-linked immunosorbent assay kits, which measured the concentration of BDNF in serum according to manufacturer’s instructions (R&D Systems, Minneapolis, MN). Samples were run in duplicate, and results were averaged.

Data Analysis

All analyses were conducted with SPSS version 23.0 (IBM Corp., Armonk, NY). Preexercise values for mood, BDNF, and depressive symptoms were analyzed with one-way ANOVA to assess potential baseline differences across sessions. Averages for HR, RPE, W, and RER across the 20-min exercise session (excluding warm-up and cooldown) were then compared among the sessions via one-way ANOVA. Each individual’s RPE for their preferred session at 0, 5, 10, 15, and 20 min after warm-up was averaged to obtain an approximation of their average intensity across the 20-min exercise bout. To examine the potential effects of antidepressant usage on mood and BDNF, a 2 (antidepressant status) × 4 (session; light, moderate, hard, and preferred) ANOVA was conducted.

Using a within-subjects design, affective and biological responses to the preferred session were compared with responses from light-, moderate-, or hard-intensity sessions that were matched to the preferred session. The responses to the preferred session were compared with two sessions on an individual basis—their closest and greatest sessions. The operational definition of the closest session was the prescribed session that had the closest RPE to the mean RPE from the preferred session. The operational definition for the greatest session was the prescribed session (either light, moderate, or hard) that resulted in the greatest change (i.e., greatest antidepressant effect) from preexercise to 30 min postexercise on the POMS Depression score.

Changes in POMS Depression scores at 10 and 30 min for each of the three sessions were compared with zero using one-sample t-tests. Changes in POMS Depression at 10 and 30 min and BDNF responses from the closest session were compared using paired-samples t-tests. The changes in POMS Depression at 10 and 30 min and BDNF responses from the greatest session were then compared with the preferred session using paired-samples t-tests.

Cohen’s d effect sizes (11) were used to further characterize differences between sessions. The level of significance was set at α = 0.05 for all statistical tests. All data are presented as mean ± SD.


Fourteen participants were currently taking a selective serotonin reuptake inhibitor or serotonin antagonist/inhibitor, three participants were also taking an additional antidepressant medication, and three were taking an anxiety medication. Ten participants were not taking any antidepressant medication. None of the participants reported the use of psychotherapy, and none made changes in medication use during the study. Results of the 2 × 4 ANOVA demonstrated that antidepressant usage did not significantly affect the POMS depression change 30 min after exercise (F3,20 = 0.964, P = 0.429). In addition, antidepressant usage did not alter the BDNF response (F3,18 = 0.123, P = 0.946). Thus, antidepressant status was not included in subsequent analyses. There was a single participant who had an external event that significantly affected her mood between 10 and 30 min after her light-intensity session that was unrelated to the study; as such, her 10-min mood ratings were carried forward to the 30-min time point. In addition, blood was not drawn after the preferred session for one participant, so she is not included in the BDNF analyses (N = 23).

Baseline and Exercise Intensity Outcomes

There were no significant differences in mood (POMS) or depression (BDI) at baseline among the four sessions (P > 0.05; Table 1). Average exercise outcomes for the sessions are presented in Table 2. For each exercise variable, the prescribed exercise sessions (light, moderate, and hard) significantly differed from one another (ANOVA; P < 0.05). The mean values for the preferred session were between those of the light and moderate intensities, except HR having a mean value between moderate- and hard-intensity sessions.

Participant characteristics and baseline mood ratings by exercise session.
Exercise variables by session.

For the preferred session, participants chose a wide range of exercise intensities ranging from 8.5 (very light) to 16 (very hard), averaging a moderate intensity (mean session average RPE = 12.5), although individuals ended the session anywhere from very light exertion (RPE = 9) to very hard exertion (RPE = 17). Some participants chose to exercise at a relatively consistent level, whereas others chose to do intervals of higher- and lower-intensity activity during the 20 min. Each individual participant’s RPE at 5-min intervals during the preferred session is illustrated in Figure 1. Each participant’s wattage across the 20 min, measured every 30 s, is illustrated in the supplemental digital content (see Figure, Supplemental Digital Content 2, Wattage across Sessions, The participants’ prescribed sessions were performed at a relatively constant wattage (with wattage decreasing over time for some in the hard-intensity condition to maintain the RPE criterion). In the preferred session, some participants maintained a consistent wattage, whereas others altered their workload throughout the 20 min.

Each participant’s RPE (measured at 5-min intervals) across the preferred intensity bout. Minute 0 corresponds to the end of the 5-min warm-up at the beginning of the 20-min preferred exertion bout. The black dotted line indicates the mean value across the participants at each of the five measurement points.

Depression and BDNF Responses to Exercise Sessions

Average changes for POMS depression scores (10 and 30 min postexercise) and BDNF for the preferred, closest, and greatest sessions are presented in Figure 2. Changes in POMS Depression scores for preferred, closest, and greatest were all significantly different from zero (all P < .004), indicating significant improvements from baseline. Characteristics for the comparison sessions are reported in Table 3 and showed that the majority of participants’ closest sessions were light (n = 9) or moderate (n = 12) with few being hard (n = 3), and this was similar to the distribution for the greatest sessions (low n = 10, moderate n = 9, and hard n = 5). Two participants’ light sessions were both their closest and greatest sessions, and three participants’ moderate sessions were their closest and greatest sessions. On an individual basis, the average difference between the comparison sessions (closest or greatest) was extremely small for each of the exercise variables (Table 3; no significant mean differences), indicating that the comparison sessions were physiologically similar to the preferred session. Change scores for the other POMS subscales and total mood disturbance followed a similar, although less pronounced, pattern after the preferred, closest, and greatest sessions (see Table, Supplemental Digital Content 3, Changes in POMS ratings for the preferred, closest, and greatest sessions,

Closest and greatest session characteristics.
Affective (A) and biological (B) responses to the exercise sessions. *Significant difference between the comparison session and the preferred session (P < 0.05 from paired-samples t-test).

Closest POMS

Depression scores at 10 min after exercise were not significantly different between the closest and the preferred sessions (closest = −5.3 ± 6.2 vs preferred = −3.4 ± 4.9, t23 = 1.946, P = 0.064, d = 0.34). Similarly, POMS Depression scores at 30 min after exercise were not significantly different between the closest and the preferred sessions (closest = −5.6 ± 6.7 vs preferred = −3.4 ± 4.8, t23 = 1.911, P = 0.069, d = 0.38).

Closest BDNF

BDNF responses were significantly greater after the closest prescribed exercise session compared with the preferred session (closest = 5.4 ± 6.9 vs preferred = −1.4 ± 9.8, t22 = −2.785, P = 0.011, d = −0.81).

Greatest POMS

Depression scores at 10 min were significantly different between the greatest and the preferred sessions (greatest = −10.8 ± 6.9 vs preferred = −3.4 ± 4.9, t23 = 5.278, P < 0.001, d = 1.25), indicating a larger mood improvement for the greatest session. Likewise, POMS depression scores at 30 min were significantly different between the greatest and the preferred sessions (greatest = −11.8 ± 7.4 vs preferred = −3.4 ± 4.8, t23 = 5.484, P < 0.001, d = 1.37), again indicating a larger mood improvement for the greatest session.

Greatest BDNF

There was not a significant difference for BDNF responses between the greatest and the preferred sessions (greatest = 1.0 ± 9.5 vs preferred = −1.4 ± 9.8, t22 = −0.824, P = 0.419, d = −0.24).


This study sought to determine whether exercise performed at a preferred intensity yielded better affective and biological benefits than exercise intensities that were prescribed. Overall, the hypotheses that preferred exercise intensity would result in greater improvements in mood and greater increases in BDNF were not supported. When comparing the preferred exercise session to the prescribed session that was performed at a similar average intensity (closest), there was a similar change in depression although a significantly smaller BDNF response. When compared with the prescribed session that resulted in the greatest mood improvement, the preferred exertion session resulted in a significantly smaller antidepressant response compared with the most effective prescribed session (greatest). These results suggest that although exercising at a preferred intensity results in acute improvements in mood, being directed to exercise at a particular intensity may be more effective. This seems to be the case both for increasing circulating BDNF and for realizing greater affective benefits. Antidepressant usage did not alter the mood or BDNF responses to exercise. Overall, the present study’s results suggest that exercising at a self-selected intensity may be less effective than prescribed exercise to promote acute improvements in mood in those with MDD.

The lack of a differential mood response between the closest and the preferred sessions suggests that patient preference does not have a large effect on the mood response. This comparison was designed to match sessions in which the participant performed a similar exercise bout with the major difference being who determined the session’s intensity (i.e., external vs internal decision). This comparison should minimize the influence of what the participant actually performed (i.e., the effects of exercise for 30 min on a cycle ergometer in a laboratory at the selected intensity) and represent the difference between the participant choosing the intensity and being prescribed a similar exercise bout. The lack of statistical difference in mood responses after these sessions suggests that preference did not play a substantial role in determining the response to the bout. However, the comparison between the two responses (d = 0.34, P = 0.064 at 10 min, d = 0.38, P = 0.069 at 30 min) showed a small effect in favor of the prescribed session, providing support to the assertion that allowing patients to self-select exercise intensity is unlikely to yield optimal improvements in depressed mood in MDD. Although speculative, it is possible that the external prescription of a goal may have resulted in better mood responses in the closest sessions although the perceptual and physiological stimuli were similar. Future research is needed to determine if and how external prescription may be more affectively beneficial than self-selection.

The stark differences between the mood responses to the preferred session and the greatest session were unexpected. This comparison was designed to consider the prescribed exercise session that led to the largest mood response for each participant and relate its effects to those after the preferred session. The initial hypothesis was that preferred exertion exercise would result in a better response than the greatest session. This hypothesis was not supported. Instead, large differences were found in favor of the greatest session at both 10 and 30 min postexercise (d = 1.25 and 1.37, respectively). This occurred despite the two conditions being similar with respect to average exercise intensity (Table 3). Moreover, light-, moderate-, and hard-intensity prescribed exercises led to the largest antidepressant response for individual participants (n = 10, 9, and 5, respectively), indicating that there was not a single optimal intensity for all participants. This variability in both the greatest and the preferred sessions from light- to hard-intensity exercise shows that the optimization of the affective response is unlikely to occur solely through intensity prescription. Indeed, recommendations for participants to self-monitor their exercise intensity to impart positive affective responses (15) may be inappropriate for a depressed population. As the affective response to a bout of exercise can predict future exercise behavior (44), these results should prompt future research to analyze how exercise sessions and programs can result in the magnitude of affective response observed after the greatest session.

Previous research has demonstrated that the self-selection or preference of exercise parameters (e.g., mode and intensity) results in improved feelings related to the exercise in healthy populations (2,3,15,22,29). Our results suggest that a different interaction between prescribed and preferred exercise intensities may exist in MDD. Although the preferred session led to improvements in depressed mood, the prescribed comparison sessions produced superior improvements (effect size differences in changes at 30 min: closest, d = 0.38 and greatest, d = 1.37). In healthy college-age women, Miller et al. (27) found improvements in positive affect after exercise of a preferred mode (choice from five aerobic machines) compared with a prescribed session, but there were similar improvements in negative affect between the sessions. Their lack of an effect of preferred exercise mode on negative affect may relate to the similar lack of a differential improvement in depressed mood (a negative affective scale) between the preferred and the closest sessions in the present study. Future research should further analyze preferred exertion exercise in healthy individuals and in other mental health conditions to determine how exercise preferences more broadly relate to affect and how this can inform exercise prescription for psychological improvement.

The present acute exercise results are in contrast to the results of preferred exertion exercise training from Callaghan et al. (8), who showed that a preferred exertion training program (12 sessions for 4 wk) provided larger improvements in BDI scores than a prescribed training program. The conflicting results regarding the effects of preferred exertion exercise in depression between Callaghan et al. and the present study may be due to the acute versus chronic exercise stimuli, but may also be due to the methods used for preferred exertion. In their training study, the preferred exertion group also received targeted motivational support based on the Transtheoretical Model of Behavior Change (31) for 15 min after each exercise session. The prescribed sessions were also intentionally designed to create a “less easy and comfortable session with unpleasant training effect accumulation and lower levels of locus control and self-efficacy”(7). The present study did not use motivational techniques, nor were the prescribed intensities meant to be necessarily less comfortable than the preferred session. Thus, the methodological differences between the two studies may explain the inconsistent findings. In addition, their most recent evaluation of preferred exertion training in depressed adolescents did not find a significant improvement in depression after an exercise program (9). Although allowing the exerciser to self-select their intensity may be useful in some contexts, the current evidence questions its utility for improving depressed mood acutely in MDD.

The differential BDNF responses across the sessions were also unexpected. Our results indicated that participating in a similar intensity, steady-state prescribed session (closest) prompted higher levels of serum BDNF than the self-selected session. Furthermore, the session associated with the greatest antidepressant response had little effect on BDNF. As BDNF may be one mechanism through which exercise training improves mood (35), the present results may indicate that training studies should prescribe exercise to improve depression rather than allow participants to choose their own intensity. It is possible that chronic exercise training may have a “compounding” effect on BDNF release that cannot be captured in an acute exercise study. It is likely that different mechanisms relate to the mood responses to acute and chronic exercise; therefore, this study does not necessarily eliminate BDNF as a potential antidepressant mechanism of exercise but rather challenges whether BDNF plays a role in the antidepressant effect of acute exercise in depression.

Others have found that acute exercise increases circulating BDNF (40), along with suggestions of a dose–response relationship (19,34). The preferred session in the present study was most often similar to the light or moderate sessions (Table 3), which could partially explain the minimal BDNF change in the preferred session (Fig. 2B). However, the closest and greatest sessions were both physiologically similar to the preferred session, suggesting that there should be similar BDNF responses if there was a strict dose–response relationship. Yet only the closest session prompted a significantly different response, being associated with a greater postexercise change in BDNF than the preferred session. Therefore, the present results suggest that the prescription of an exercise stimulus may increase the BDNF response (closest) and that an individually optimized acute exercise bout does not result in greater-than-normal BDNF (greatest). Similar to the comparison sessions here, nontraditional exercise prescriptions may alter the theoretical dose–response relationship of BDNF. This highlights the utility of considering nonphysiological relationships between exercise and mood (e.g., neurobiological influence of exercise preference) and what they can teach us about the psychobiology of exercise.

At the end of each participant’s final visit, there was a debriefing session where participants were questioned about their reactions to the preferred versus prescribed sessions and allowed to ask any questions and discuss any part of the study. Multiple participants reported that as the preferred session lacked a goal, they did not feel that the exercise was as beneficial to them as the prescribed sessions. The lack of a goal could have been an important moderator of the smaller mood improvement following this condition. For example, one participant reflected that she felt that no matter what she chose to do on that day, it was not going to be effective because of her own feelings of low self-worth and self-efficacy. This is an important consideration and suggests that exercise prescription for patients with MDD may benefit from realistic goal setting and occurring in settings where they can reasonably anticipate a positive effect on their mood.

This study has several limitations. We chose a single mode of exercise (cycling) to test our hypotheses, thus allowing us to selectively manipulate one aspect of exercise preference (intensity). It is possible that allowing a participant to choose their desired mode of exercise (e.g., resistance training, jogging, etc.) could have affected the antidepressant results of the preferred session (27). There was a small sample size in the present study with 24 participants. However, significant differences were seen between the preferred session and the prescribed sessions demonstrating sufficient power to find differential effects. In addition, the psychobiological differences between the exercise sessions could be due to the greater intensity modifications in the preferred session (Fig. 1) compared with steady-state exercise in the prescribed sessions. Although this is possible, the intention of the preferred session was to test whether instructing participants to exercise however they preferred without putting any intensity limitations on them would result in a greater antidepressant response than a prescribed steady-state session (i.e., a traditional prescription). Because of the nature of the question, allowing participants to alter intensity continuously better mimicked an exercise session that they might engage in on their own.


Exercising at a preferred intensity led to smaller acute improvements in mood compared with prescribed steady-state exercise bouts in women with MDD. Exercise intensity preference was associated with a smaller BDNF response than the response after a similar-intensity prescribed bout. Overall, prescribed exercise undertaken to improve mood in MDD may lead to greater outcomes (psychologically and biologically) than allowing the patient to choose. Clinicians, psychologists, and other practitioners should consider providing clear exercise intensity recommendations for symptom management in depression rather than allowing patients to self-select their intensity. There was no consistent difference between the preferred session and the session that provided the greatest improvement in depressed mood, making it difficult to provide a single, generalizable prescription that would lead to an optimal effect across all individuals. Further examination of the psychobiological consequences of exercise will help to determine prescriptions that maximize exercise’s mood-enhancing effects.

This project was funded, in part, by the Virginia Horne Henry Gift Fund, the University of Wisconsin–Madison Graduate School, and the Wisconsin Center for Education Research. Jacob Meyer was partially supported by a National Research Service Award from the Health Resources and Services Administration T32HP10010 to the University of Wisconsin Department of Family Medicine and Community Health. None of the funding sources were involved in the study design, collection, analysis, or interpretation of the data. The authors would like to thank Lauren Schlapman, Hannah Feinstein, Shawn Tipple, Matthew Patton, Caroline Wickler, and Rachel Prince for their assistance. The authors report no conflicts of interest. Results of this study do not constitute endorsement by the American College of Sports Medicine.


1. Beck AT, Steer RA, Ball R, Ranieri W. Comparison of Beck Depression Inventories-IA and -II in psychiatric outpatients. J Pers Assess. 1996;67(3):588–97.
2. Berger BG, Darby LA, Zhang Y, Owen DR, Tobar DA. Mood alteration after 15 minutes of preferred intensity exercise: examining heart rate, perceived exertion, and enjoyment. J Sport Behav. 2016;39(1):3–21.
3. Berger BG, Motl RW. Exercise and mood: a selective review and synthesis of research employing the Profile of Mood States. J Appl Sport Psychol. 2000;12(1):69–92.
4. Borg G. Physical performance and perceived exertion [Internet]. 1962; [cited 2016 Mar 28] Available from:
5. Bromet E, Andrade LH, Hwang I, et al. Cross-national epidemiology of DSM-IV major depressive episode. BMC Med. 2011;9:90.
6. Brunoni AR, Lopes M, Fregni F. A systematic review and meta-analysis of clinical studies on major depression and BDNF levels: implications for the role of neuroplasticity in depression. Int J Neuropsychopharmacol. 2008;11(8):1169–80.
7. Callaghan P, Khalil E, Morres I. Pragmatic Randomised Controlled Trial of a Preferred Intensity Exercise Programme to Improve Wellbeing Outcomes of Women Living with Depression. University of Nottingham; 2009. [cited 2016 Jan 29] Available from:
8. Callaghan P, Khalil E, Morres I, Carter T. Pragmatic randomised controlled trial of preferred intensity exercise in women living with depression. BMC Public Health. 2011;11:465.
9. Carter T, Morres I, Repper J, Callaghan P. Exercise for adolescents with depression: valued aspects and perceived change. J Psychiatr Ment Health Nurs. 2016;23(1):37–44.
10. Coelho FG, Gobbi S, Andreatto CA, Corazza DI, Pedroso RV, Santos-Galduróz RF. Physical exercise modulates peripheral levels of brain-derived neurotrophic factor (BDNF): a systematic review of experimental studies in the elderly. Arch Gerontol Geriatr. 2013;56(1):10–5.
11. Cohen J. Statistical Power Analysis for the Behavioral Sciences. 1st ed. New York and London: Academic Press; 1969. pp. 415.
12. Craig CL, Marshall AL, Sjöström M, et al. International physical activity questionnaire: 12-country reliability and validity. Med Sci Sports Exerc. 2003;35(8):1381–95.
13. Dishman RK. Prescribing exercise intensity for healthy adults using perceived exertion. Med Sci Sports Exerc. 1994;26(9):1087–94.
14. Dunn AL, Trivedi MH, Kampert JB, Clark CG, Chambliss HO. Exercise treatment for depression: efficacy and dose response. Am J Prev Med. 2005;28(1):1–8.
15. Ekkekakis P, Parfitt G, Petruzzello SJ. The pleasure and displeasure people feel when they exercise at different intensities: decennial update and progress towards a tripartite rationale for exercise intensity prescription. Sports Med. 2011;41(8):641–71.
16. Eston RG, Lamb KL, Parfitt G, King N. The validity of predicting maximal oxygen uptake from a perceptually-regulated graded exercise test. Eur J Appl Physiol. 2005;94(3):221–7.
17. Eston RG, Williams JG. Reliability of ratings of perceived effort regulation of exercise intensity. Br J Sports Med. 1988;22(4):153–5.
18. Faulkner J, Parfitt G, Eston R. Prediction of maximal oxygen uptake from the ratings of perceived exertion and heart rate during a perceptually-regulated sub-maximal exercise test in active and sedentary participants. Eur J Appl Physiol. 2007;101(3):397–407.
19. Ferris LT, Williams JS, Shen CL. The effect of acute exercise on serum brain-derived neurotrophic factor levels and cognitive function. Med Sci Sports Exerc. 2007;39(4):728–34.
20. Garber CE, Blissmer B, Deschenes MR, et al. American College of Sports Medicine Position Stand: quantity and quality of exercise for developing and maintaining cardiorespiratory, musculoskeletal, and neuromotor fitness in apparently healthy adults: guidance for prescribing exercise. Med Sci Sports Exerc. 2011;43(7):1334–59.
21. Glass SC, Knowlton RG, Becque MD. Accuracy of RPE from graded exercise to establish exercise training intensity. Med Sci Sports Exerc. 1992;24(11):1303–7.
22. Hamlyn-Williams CC, Freeman P, Parfitt G. Acute affective responses to prescribed and self-selected exercise sessions in adolescent girls: an observational study. BMC Sports Sci Med Rehabil. 2014;6:35.
23. Julian LJ. Measures of anxiety: State-Trait Anxiety Inventory (STAI), Beck Anxiety Inventory (BAI), and Hospital Anxiety and Depression Scale-Anxiety (HADS-A). Arthritis Care Res. 2011;63(11 Suppl):S467–72.
24. Matrisciano F, Bonaccorso S, Ricciardi A, et al. Changes in BDNF serum levels in patients with major depression disorder (MDD) after 6 months treatment with sertraline, escitalopram, or venlafaxine. J Psychiatr Res. 2009;43(3):247–54.
25. McNair DM, Lorr M, Droppleman LF. Profile of Mood States: Manual. San Diego (CA): Educational and Industrial Testing Services (Edits); 1992.
26. Meyer JD, Koltyn KF, Stegner AJ, Kim J-S, Cook DB. Influence of exercise intensity for improving depressed mood in depression: a dose–response study [Internet]. Behav Ther. [cited 2016 Apr 28] Available from:
27. Miller BM, Bartholomew JB, Springer BA. Post-exercise affect: the effect of mode preference. J Appl Sport Psychol. 2005;17(4):263–72.
28. Paluska SA, Schwenk TL. Physical activity and mental health: current concepts. Sports Med. 2000;29(3):167–80.
29. Parfitt G, Rose EA, Markland D. The effect of prescribed and preferred intensity exercise on psychological affect and the influence of baseline measures of affect. J Health Psychol. 2000;5(2):231–40.
30. Piepmeier AT, Etnier JL. Brain-derived neurotrophic factor (BDNF) as a potential mechanism of the effects of acute exercise on cognitive performance. J Sport Health Sci. 2015;4(1):14–23.
31. Prochaska JO, Velicer WF. The transtheoretical model of health behavior change. Am J Health Promot. 1997;12(1):38–48.
32. Reddon JR, Marceau R, Holden RR. A confirmatory evaluation of the Profile of Mood States: convergent and discriminant item validity. J Psychopathol Behav Assess. 1985;7(3):243–59.
33. Rethorst CD, Trivedi MH. Evidence-based recommendations for the prescription of exercise for major depressive disorder. J Psychiatr Pract. 2013;19(3):204–12.
34. Rojas Vega S, Strüder HK, Vera Wahrmann B, Schmidt A, Bloch W, Hollmann W. Acute BDNF and cortisol response to low intensity exercise and following ramp incremental exercise to exhaustion in humans. Brain Res. 2006;1121(1):59–65.
35. Sartori CR, Vieira AS, Ferrari EM, Langone F, Tongiorgi E, Parada CA. The antidepressive effect of the physical exercise correlates with increased levels of mature BDNF, and proBDNF proteolytic cleavage-related genes, p11 and tPA. Neuroscience. 2011;180:9–18.
36. Schmidt HD, Duman RS. Peripheral BDNF Produces Antidepressant-Like Effects in Cellular and Behavioral Models. Neuropsychopharmacology. 2010;35(12):2378–91.
37. Schuch FB, Vancampfort D, Richards J, Rosenbaum S, Ward PB, Stubbs B. Exercise as a treatment for depression: a meta-analysis adjusting for publication bias. J Psychiatr Res. 2016;77:42–51.
38. Sheehan DV, Lecrubier Y, Sheehan KH, et al. The Mini-International Neuropsychiatric Interview (M.I.N.I.): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J Clin Psychiatry. 1998;59(20 Suppl):22–33;quiz 34–57.
39. Spielberger C. Manual for the State-Trait Anxiety Inventory. Palo Alto (CA): Counseling Psychologists Press; 1983. p. 36.
40. Szuhany KL, Bugatti M, Otto MW. A meta-analytic review of the effects of exercise on brain-derived neurotrophic factor. J Psychiatr Res. 2015;60:56–64.
41. Thomas S, Reading J, Shephard RJ. Revision of the Physical Activity Readiness Questionnaire (PAR-Q). Can J Sport Sci. 1992;17(4):338–45.
42. Trivedi MH, Greer TL, Church TS, et al. Exercise as an augmentation treatment for nonremitted major depressive disorder: a randomized, parallel dose comparison. J Clin Psychiatry. 2011;72(5):677–84.
43. Williams DM, Anderson ES, Winett RA. A review of the outcome expectancy construct in physical activity research. Ann Behav Med. 2005;29(1):70–9.
44. Williams DM, Dunsiger S, Ciccolo JT, Lewis BA, Albrecht AE, Marcus BH. Acute affective response to a moderate-intensity exercise stimulus predicts physical activity participation 6 and 12 months later. Psychol Sport Exerc. 2008;9(3):231–45.


Supplemental Digital Content

© 2016 American College of Sports Medicine