Table 2.

Summary of relationship of cardiorespiratory fitness and motor skills with memory functions

Reference, country	Design/subjects	Fitness or motor skill	Memory	Main results
Chaddock et al. (2010), USA	Cross-sectional 59 children 9–10 years of age	Maximal treadmill testa	Item and relational memory paradigm + MRI	HF had greater bilateral hippocampal volumes and better relational memory than UF; bilateral hippocampal volume mediated the relationship between fitness and relational memory. No association between fitness and item memory was found
Kamijo et al. (2011), USA	RCT (9 months) 36 children 7–9 years of age	Maximal treadmill exercise test to assess VO2max before and after exercise intervention	Modified Stenberg task for item memory + ERP	Increased fitness after the intervention was associated with improved overall accuracy in memory task. The intervention group exhibited greater iCNV amplitude in comparison to control group at the end of the intervention period
Monti et al. (2012), USA	Longitudinal (9 months) 44 children 9 years of age, cognitive assessment of only in the follow-up	Maximal treadmill exercise test to assess VO2max before and after exercise intervention	Working memory task for non-relational and relational memory, eye-movement tracking	Children in the exercise intervention group showed longer viewing times for the related pictures than control group during relational memory task. No differences during non-relational memory task accuracy. No differences in performance accuracy or reaction times between intervention and control groups
Niederer et al. (2011), Switzerland, Germany	Cross-sectional /longitudinal (nine-month follow-up) 245 children 5.2 years of age	Agility, dynamic balance, multistage shuttle run test	Item memory	Agility was associated with item memory in cross-sectional analyses and baseline balance predicted improvement in item memory longitudinally
Pangelinan et al.(2011), USA	Cross-sectional 172 children 6– 13 years of age	Purdue pegboard task	Spatial working memory + MRI	Pegboard test performance was not related to spatial working memory.
Piek et al. (2008), Australia	Cohort study 33 children 6–11 years of age at the follow-up (initial assessment at the age of four month)	Ages and stages questionnaire, McCarron assessment of neuromuscular development	Digit span and letter-number sequencing = working memory index	Early gross motor development predicted working memory
Roebers and Kauer (2009), Switzerland	Cross-sectional 112 children 7.5 years of age	Sideway jumping, moving sideways, postural flexibility, pegboard	Backward colour recall	Better performance in the postural flexibility test was related to higher performance in memory task
Wassenberg et al.(2005), the Netherlands	Cross-sectional 378 children 6.18 years of age	Maastricht motor test (static balance, dynamic balance, ball skills, manual dexterity)	Auditory item memory	Better qualitative motor skills (i.e. the movement process) were related to a better item memory

HF, highly fit children. UF, unfit children. ERP, event-related brain potential.

iCNV, initial contingency negative variation

^aDivided into > 70^th and < 30^th percentile according to child’s VO_2max.