Table 4 Summary of evidence of the seven articles included in the review

Blain et al. [14]

37 male triathletes:

n = 19 FOR (35 ± 1.2 years)

n = 18 control (36 ± 1.5 years)

8-week study design with 3-week overload protocol (40% ↑ in training load)

Behavioural choice task (i.e. subjective preference) and a cognitive task (i.e. N-Back/N-Switch)

Training induced fatigue ↑ the attraction of immediate versus delayed rewards in economic choice task; but not the way value options were estimated and compared. No sig. differences between groups in cognitive task performance

FOR

Dupuy et al. [15]

10 male endurance athletes (31 ± 6 years):

n = 6 road running

n = 2 road cycling

n = 2 triathletes

2-week overload period (100% ↑ from baseline training volume)

5 athletes were assigned to the negative adaptation group (NAG) following overload training

Simple reaction time task and a computerised version of the Stroop colour-word test

No sig. differences between groups in Stroop task performance. Effect size show small ↑ in NAG and a moderate ↓ in PAG in reaction time. Error rate did not change in congruent, denomination or interference tasks. For switching task ↓ after overload training in NAG, suggesting a ↓ in performance that required the use of executive function with FOR

Sig. group × time interaction for simple reaction time, with initiation time significantly slower for NAG following overload

FOR

Dupuy et al. [16]

11 male endurance athletes (29.5 ± 9.3 years):

n = 6 road running

n = 2 road cycling

n = 3 triathletes

2-week overload period (100% ↑ from baseline training volume)

Computerised modified Stroop Test

Slowing of cognitive performance following 2-week overload period. Moderate ↑ in overall RT to Stroop tasks after overload (p = 0.03) returned to base after 1 week taper, with small/moderate ↑ in RT for denomination (p = 0.04) and interference conditions (p = 0.01) after overload. NS tendency for increase in FT with switching condition (p = 0.07)

FOR

Hynynen et al. [13]

24 endurance athletes:

n = 12 Control

n = 12 overreaching

Both groups 50:50 male: female split

Clinically diagnosed as OTS versus control group

Stroop colour word test

Overtrained athletes made sig. more errors than control, suggesting signs of ↓ cognitive performance when overtrained compared to a normal training state. OTS athletes performed poorer than control when time to respond to stroop task was reduced (moderate, p = 0.046; and fast speeds, p = 0.045). No difference (similar) in slow speed

OTS

Le Meur et al. [18]

24 triathletes:

n = 8 normal training

n = 16 intense trained (IT)

Sex not reported

IT group completed a 3-week intensified programme designed to deliberately overtrain the triathletes; Duration of each training session of the classic training period was ↑ by 40%

*Only 11 of 16 were truly OR

Audio stimulus reaction time test

No difference in cognitive performance at low intensity function between control group and overtraining group; however, at exhaustion there was a sig. ↓ in cognitive performance

OR group showed sig. ↓ in performance (more false responses to audio stimulus task) only at exhaustion (not at rest, low intensity, or lactate threshold) relative to normal training group

FOR

Nederhof et al. [19]

28 cyclists:

n = 14 control

(9 male; 5 female)

n = 14 overreaching

(10 male; 4 female)

of which: n = 7 well trained; n = 5 FOR; n = 2 excluded from analysis

The high load training period consisted of a regular training camp during which the cyclists performed their own training schedule. The training camp lasted on average 9.5 days (SD = 2.8). This was for the 14 well trained cyclists

Psychomotor speed: finger pre-cueing task

Determination test

No sig. differences in psychomotor speed between groups on finger pre-cueing task; trend for reduced psychomotor speed in FOR athletes after training overload versus control. Again, on determination task, trend for delayed in reaction time for the two faster presentation intervals for participants after the training camp group versus control. No group differences for action part of determination test; no group differences for reaction times of determination test

FOR

Rietjens et al. [17]

14 male cyclists

n = 7 experimental group (25.3 ± 4.7 years)

n = 7 age matched control group

2-week intensified training period, preceded by 2-week pre-intervention baseline training period

Video stimuli reaction time test

The most sensitive parameter for detecting overtraining was reaction time

FOR/NFOR unable to distinguish as no measures after recovery period