Skip to main content

Table 4 Motor imagery assessments: The characteristics of the included studies - Validity

From: Imagery ability assessments: a cross-disciplinary systematic review and quality evaluation of psychometric properties

Tool

Disciplines

Study

Country

Language

Study population

Validity

COSMIN

Quality criteria

Comments

Participants

N

Age mean (years)

Sex

Design

Results

Kinesthetic and Visual Imagery Questionnaire (KVIQ)

Med

Malouin et al. 2007 [43]

CA

E

Strokea

33

60.1

7♀, 26♂

Construct validity- structural validity

KVIQ-20 + KVIQ-10

PCA and oblique rotation extracted two factors for both versions.

Correlation between the two factors for both versions was 0.46. Factor loadings for KVIQ-20 ranged from 0.70 to 0.88 (visual) and 0.68 to 0.80 (kinaesthetic); for KVIQ-10 ranged from 0.73 to 0.86 (visual) and 0.68 to 0.80 (kinaesthetic). Total variance explained by 63.4% for KVIQ-20 and 67.7% for KVIQ-10

Adequate

+

EFA applied, factors loading >0.40, variance explained less than 50%, corr. among factors reported.

Healthyb

70

42.9

49♀, 21♂

LL amputationc

13

35.0

13♂

Acquired blindnessd

10

40.8

4♀, 6♂

LL immobilizatione

5

50.1

5♂

Med

Randhawa et al. 2010 [68]

CA

E

PD

11

61.7

7♀, 4♂

Construct validity- hypothesis testing

Corr. KVIQ-20 and MIQ-R

r=0.94 kinaesthetic

r=0.88 visual

r=0.93 for total score

Inadequate

+

Sample size included in this analysis not adequate.

Strong corr. with instruments measuring the same construct.

Med

Schuster et al. 2012 [67]

CH

G

Subacute stroke

Chronic stroke

Left parietal lobe

MS

PD

19

59.9

6♀, 13♂

Construct validity- hypothesis testing

Corr. KVIQ-G and Imaprax-G

r=0.36 visual (KVIQ-G-20 vs. Imaprax)

r=0.32 visual (KVIQ-G-10 vs. Imaprax)

Doubtful

−

Small sample size.

Only patients, who chose the internal perspective, were analysed.

Low corr. with instruments measuring the same construct.

73

62.8

28♀, 45♂

Construct validity- structural validity

KVIQ-G-20

PCA and promax rotation identified bifactorial structure of the KVIQ-G-20. Factor loadings for kinaesthetic subscale 0.79–0.93 and 0.68–0.91 for visual. Total variance of both factors explained by 69.7%

Inadequate

?

EFA applied, factors loading >0.40, variance explained less than 50%, corr. among factors reported but very low sample size.

Med

Tabrizi et al. 2013 [69]

IR

NR

MS

15

31.7

12♀, 3♂

Construct validity- hypothesis testing

Corr. KVIQ-20 and MIQ

r=0.75 kinaesthetic

r=0.78 visual

Doubtful

+

*Insufficient information about factor analysis reported for quality criteria rating.

Strong corr. with instruments measuring the same construct.

Construct validity- structural validity

KVIQ-20

Bifactorial structure of the KVIQ-20 was confirmed. Total variance of both factors explained by 90%

Inadequate

?

Med

Nakano et al. 2018 [71]

JP

J

Students

28

20.6

13♀, 15♂

Construct validity- hypothesis testing

Corr. KVIQ-20 and MIQ-R

r=0.77 kinaesthetic

r=0.64 visual

Corr. KVIQ-10 and MIQ-R

r=0.78 kinaesthetic

r=0.62 visual

Doubtful

+

Sample size calculation not mentioned.

Small sample size.

Strong corr. with instruments measuring the same construct.

Movement Imagery Questionnaire (MIQ)

Sport

Hall et al. 1985 [72]

CA

E

Students

80

NR

NR

Construct validity- stability of the internal structure

Corr. kinaesthetic vs. visual subscale

Correlation between the score achieved on the both subscales (kinaesthetic and visual) was 0.58

NA

NA

Factor structure was not analysed.

Only the total score corr. for both subscales was reported and authors suggest the stability of the subscale structure.

n.d.s

Atienza & Balaguer 1994 [73]

ES

E

Students

110

20.1

47♀, 63♂

Construct validity- structural validity

Common factor analysis using maximum likelihood and oblique rotation confirmed extracted two factors. Factor loadings for visual ranged from 0.58 to 0.82 and for kinaesthetic 0.46 to 0.81. Total variance explained by 47.8%.

adequate

?

Explained variance <50%, but all factors loaded >0.40. Corr. among factors not reported.

n.d.s

Lequerica et al. 2002 [22]

USA

E

Students

80

22.1

41♀, 39♂

Construct validity- hypothesis testing

Corr. MIQ and GTVIC

r=0.45 MIQ visual

Corr. MIQ and VMIQ

r=0.56 kinaesthetic; r=0.52 visual

Doubtful

+

#, Insufficient information on measurement properties of the comparator measures. The results in accordance with hypothesis: sign. corr. among subjective measures of mental imagery.

No corr. between subjective and objective measures of mental imagery ability providing the evidence for the multidimensional nature of imagery.

Revised Movement Imagery Questionnaire (MIQ-R)

Psy

Hall & Martin 1997 [91]

CA

E

Students

50

20.9

26♀, 24♂

Criterion validity

Corr. MIQ and MIQ-R

r=0.77 kinaesthetic

r=0.77 visual

Doubtful

+

#, Doubtful sample size.

Corr. with gold standard- MIQ was >0.70.

Sport

Monsma et al. 2009 [74]

USA

E

Athletes and dancers

325

20.2

189♀, 136♂

Construct validity- structural validity

CFA include a path between two factors (kinaesthetic and visual) and suggest these two factors are interrelated.

∆χ2(1)=126.14, p<0.001. CFI=0.99, NNFI=0.98, AGFI=0.95, SRMR=0.03, RMSEA=0.06.

Very good

+

Accepted model fit: CFI, NNFI or AGFI >0.95, or SRMR <0.08, or RMSEA <0.06.

Sport

Williams et al. 20121 [31]

CA

E

Athletes and dancers

400

20.8

219♀, 181♂

Construct validity- structural validity

MTMM approach to CFA and two models CT and CTCU were tested. Factor loadings for both models ranged from 0.70- 0.84. Corr. between the two factors (kinaesthetic and visual) for the CT was 0.25 and for the CTCU 0.23. CTCU model provided a significantly better fit to the data compared with the CT model. χ2=25.99, df=15, CFI=0.99, TLI=0.99, SRMR=0.03, RMSEA=0.05. The kinaesthetic and visual imagery are separate but related constructs.

Very good

+

Accepted model fit: CFI, TLI>0.95, or SRMR <0.08, or RMSEA<0.06.

Movement Imagery Questionnaire- Revised second version (MIQ-RS)

Sport

Gregg et al. 2010 [75]

UK

E

Athletes

321

23.3

174♀, 146♂

Construct validity- structural validity

CFA confirmed the bifactorial (kinaesthetic and visual) structure of MIQ-RS. χ2//df=3.72, CFI=0.99, RFI=0.98, RMSEA=0.09.

Inadequate

-

MIQ-RS developed for patients with motor impairments but tested with athletes.

Should be tested in another field.

RMSEA not acceptable. SRMR not reported.

Criterion validity

Corr. MIQ-RS and MIQ-R

r=0.80 kinaesthetic

r=0.82 visual

Very good

+

Corr. with gold standard- MIQ-R was >0.70.

Med

Butler et al. 2012 [76]

USA

E

Strokea

23

59.2

7♀, 16♂

Construct validity- structural validity

PCA and varimax rotation extracted two factors: kinaesthetic and visual. Communalities ranged from 0.72 to 0.95 in the stroke and 0.72 to 0.96 in the healthy group. Corr. between the two factors (kinaesthetic and visual) in the stroke was 0.61 and in the healthy 0.69. Total variance in the stroke group was explained by 83.4% and in the healthy group by 88.6%.

Inadequate

?

All criteria for EFA fulfilled but very low sample size.

Healthyb

23

51

11♀, 12♂

Construct validity- hypothesis testing

Corr. MIQ-RS and KVIQ-10

kinaesthetic r=0.84a/ r=0.86b

visual r=0.62a/ 0.77b

Very good

+

Strong corr. with instruments measuring the same construct.

n.d.s.

Loison et al. 2013 [77]

FR

F

Healthy

153

37.9

118♀, 35♂

Construct validity- structural validity

CFA confirmed the bifactorial (kinaesthetic and visual) structure of MIQ-RS French version. Corr. between items were strong, for the kinesthetic 0.74–0.85 and for visual 0.65–0.79. Total variance explained by 55–73% for kinesthetic and 42–62% for visual. χ2//df=2.23, CFI=0.93, SRMR=0.06, RMSEA=0.09.

Very good

−

Accepted model fit: CFI >0.95, or SRMR <0.08, or RMSEA <0.06.

Movement Imagery Questionnaire-3 (MIQ-3)

Sport

Williams et al. 20122 [31]

CA

E

Athletes and dancers

370

20.3

185♀, 185♂

Construct validity- structural validity

MTMM approach to CFA and two models CT and CTCU were tested. Factor loadings for the CT model ranged from 0.70 to 0.81 and for the CTCU model ranged from 0.64 to 0.81. Corr. between the factors (external, internal and kinesthetic) for the CT was 0.33 to 0.68 and for the CTCU 0.32 to 0.60. The three-factor CTCU model provided the best fit to the data compared with the CT model: χ2=75.12, df=39, CFI=0.98, TLI=0.97, SRMR=0.04, RMSEA=0.05. The corr. between kinaesthetic and internal was strong (r = 0.60)

Very good

+

Accepted model fit: CFI or TLI >0.95, or SRMR <0.08, or RMSEA <0.06

The MIQ-3 factor structure was not invariant across gender.

Criterion validity- concurrent validity

Corr. MIQ-3 and VMIQ-2

r=0.68 external

r=0.63 internal

r=0.71 kinaesthetic

Very good

-

Corr. between MIQ-3 and VMIQ-2 only for kinaesthetic just above 0.70.

Sport

Williams et al. 20123 [31]

CA

E

Athletes

97

19.5

58♀, 39♂

Criterion validity- Predictive validity

MIQ-3 external sign. predict skill observational learning (OL) β=0.39, t=2.82, p=0.006

MIQ-3 external sign. predict strategy (OL) β=0.44, t=3.17, p=0.002

MIQ-3 kinaesthetic sign. predict performance (OL) β=0.48, t=3.30, p=0.001

Doubtful

?

Multiple regressions conducted to assess the predictive validity.

Sample size doubtful. Doubtful if FOLO could be used as external criterion.

Sport

Budnik-Przybylska et al. 2016 [78]

PL

PO

Athletes

276

21.3

102♀, 174♂

Construct validity- structural validity

CFA with maximum likehood estimation confirmed the three-factor (external, internal and kinaesthetic) structure. χ2=76.98, df=51, CFI=0.93, GFI=0.89, AGFI=0.83, RMR=0.25, RMSEA=0.04

Very good

+

Accepted model fit: CFI, GFI >0.95, or SRMR <0.08, or RMSEA <0.06.

n.d.s.

Paravlic et al. 2018 [79]

Sl

SL

Healthy

86

35.3

41♀, 45♂

Construct validity- structural validity

CFA and three-factor model achieved best model fits: χ2=75.40, df=51, CFI=0.94, TLI=0.93, RMR/SRMR=0.11, RMSEA=0.07

Adequate

−

Accepted model fit: CFI or TLI >0.95, or SRMR <0.08, or RMSEA <0.06.

Above mentioned criteria for good properties not met.

n.d.s.

Dilek et al. 2020 [80]

TR

Tu

Healthy

181

21.6

53♀, 132♂

Construct validity- structural validity

CFA and the three-factor structures previously proposed in the literature were tested using the LISREL structural equation-modelling programme developed. χ2 =115.60, df =51, P=0.000). CFI=0.97, GFI=0.91, AGFI=0.86, RMR=0.04, RMSEA=0.08, SRMR=0.05 Factor loadings 0.54–0.76.

Very good

+

Accepted model fit: CFI or GFI >0.95, or SRMR <0.08, or RMSEA <0.06.

Sport

Robin et al. 2020 [81]

FR

F

Students

172

20.2

115♀

Construct validity- structural validity

EFA identified three factors: external, internal and kinaesthetic. Explained variance by factor 1=48.63%, factor 2=14.56%, factor 3=17.71%. Factor loadings 0.74–0.92. CFA with maximum likelihood was performed: χ2=120.75, df=54, CFI=0.91, RMSR=0.07 and 0.08, RMSEA=0.09.

Very good

−

Accepted model fit: CFI or GFI >0.95, or SRMR <0.08, or RMSEA <0.06.

19.9

57♂

n.d.s.

Trapero-Asenjo et al. 2021 [82]

ES

S

Students

140

21.5

47♀, 93♂

Construct validity- structural validity

CFA and the three-factor model showed good fit: RMSEA=0.07, NFI=0.90, RFI=0.91, CFI=0.90. The absolute fit measures with χ2 of p=0.001 indicating an inadequate model.

Doubtful

-

Accepted model fit: CFI or GFI >0.95, or SRMR <0.08, or RMSEA <0.06.

Rotation method by CFA not described.

Construct validity- hypothesis testing

Corr. MIQ-3 and MIQ-R

Total score Spearmen’s r=0.89

External and visual r=0.72

Internal and visual r=0.70

Kinaesthetic scales r=0.89

Inadequate

+

No information on the measurement properties of the comparator instrument. Strong corr. with instruments measuring the same construct.

Movement Imagery Questionnaire for Children (MIQ-C)

n.d.s.

Martini et al. 2016 [83]

CA

E

Healthy children

204

9.6

125♀, 79♂

Construct validity- structural validity

MTMM approach to CFA and four models were tested. Factor loadings for the CT model ranged from 0.51 to 0.67 and for the CTCU model ranged from 0.51 to 0.69.

Corr. between the factors (external, internal and kinaesthetic) for the CT was 0.42 to 0.65 and for the CTCU 0.39 to 0.63. The three-factor CTCU model provided the best fit to the data compared with the CT model: χ2=75.33, df=39, CFI=0.93, TLI=0.89, SRMR=0.05, RMSEA=0.07.

Very good

+

Accepted model fit: CFI or TLI >0.95, or SRMR <0.08, or RMSEA <0.06.

Test of Ability in Movement Imagery (TAMI)

Psy

Madan & Singhal, 20132 [84]

CA

E

Students

49

19.6

29♀, 20♂

Construct validity- structural validity

PCA and varimax rotation confirmed that factor objective movement imagery was loaded by TAMI with 0.81.

Inadequate

?

#, EFA performed but not explicit to explore the structural validity of TAMI. *Insufficient information reported for quality criteria rating.

Construct validity- hypothesis testing

Corr. TAMI and VMIQ-2 internal visual: r=0.36, p<0.05

Corr. TAMI and FPIQ three subscales: r=0.451, r=0.392, r=0.343, p<0.05

Corr. TAMI and VVIQ:r=0.43, p<0.01

TAMI do not correlate with VMIQ-2 external and kinaesthetic subscales, with the MRT, and with the FPIQ kinaesthetic

Inadequate

?

The subscales of FPIQ: 1= position, 2= action, 3= object

No hypothesis defined. Insufficient information about measurement properties of the comparator instrument.

Psy

Madan & Singhal, 2014 [92]

CA

E

Students

189

19.5

125♀, 64♂

Construct validity- hypothesis testing

Corr. TAMIw *and VMIQ-2 internal visual: r=0.37

Corr. TAMIw and FPIQ subscale position: r=0.44

Corr. TAMIw and VVIQ: r=0.32 TAMIw does not correlate with VMIQ-2 external and kinaesthetic subscales, with the MRT, and with the FPIQ action, object and kinaesthetic subscales

Inadequate

?

#, No hypothesis defined. No information about measurement properties of the comparator instrument.

Test of Ability in Movement Imagery with Hands (TAMI-H)

Psy

Donoff et al. 2017 [93]

CA

E

Students

70

NR

49♀, 21♂

Construct validity- hypothesis testing

Corr. TAMI-H and TAMIw:

r=0.29 FM/ r=0.53 IM

Corr. TAMI-H and FPIQ kinaesthetic:

r=0.34 FM/ r=0.26 IM

Corr. TAMI-H and FPIQ position:

r=0.19 FM/ r=0.26 IM

Corr. TAMI-H and FPIQ action:

r=0.21 FM/ r=0.34 IM

Corr. TAMI-H and FPIQ object:

r=0.35 FM/ r=0.44

inadequate

?

Author mentioned that new Tool-TAMI-H (with two imagery type: Functionally-involved Movement (FM) and Isolated Movement (IM)) was developed but no information reported about development.

Measurement properties of the comparator instrument not mentioned.

Vividness of Haptic Movement Imagery Questionnaire (VHMIQ)

n.d.s.

Campos et al. 1998 [85]

ES

S

Students

338

20.9

51♀, 287♂

Construct validity- hypothesis testing

Corr. VHMIQ and VMIQ

Pearson r=0.56 for women, r=0.66 for men 0.66 and r=0.60 for all participants.

Inadequate

?

Strong corr. was expected. Not reported if different corr. between VHMIQ and internal VMIQ or VHMIQ and external VMIQ was found.

No information about measurement properties of the comparator instrument.

Construct validity- hypothesis testing

Known-groups validity

Mixes-model analysis of variance with the factor sex and type of image: neither sex (F: 2.12 p>0.05) or type of image (F: 3.24, p>0.05) had a sig. effect on reported vividness of imagery.

Doubtful

?

Results are in accordance with the hypothesis that no sex difference should be expected but no adequate description provided of important characteristics of the subgroups.

Vividness of Movement Imagery Questionnaire (VMIQ)

Sport

Isaac et al. 1986 [27]

NZ

E

Studentsa

220

NR

NR

Construct validity- hypothesis testing

Corr. VMIQ and VVIQ

Pearson corr. coefficient for group a r=0.81

Spearman rank for group b r=0.75, group c r=0.45 and group d r=0.65

Inadequate

?

Small sample size in group b, c and d.

Corr. ranged from low to strong among different groups. But group differences not reported. Insufficient information about measurement properties of the comparator instrument.

No trampoline experienceb

25

Trampoline experiencec

25

International level trampolinistsd

16

Sport

Eton et al. 1998 [86]

USA

E

Varsity athletes

51

NR

27♀, 24♂

Construct validity- hypothesis testing

Corr. VMIQ and VVIQ

r=0.60, p<0.01

Doubtful

?

Doubtful if constructs measured by comparator instrument are same.

Some information about measurement properties of the comparator instrument.

Recreational athletes

48

24♀, 24♂

Non-athletes

26

14♀, 12♂

n.d.s

Lequerica et al. 2002 [22]

USA

E

Students

80

22.1

41♀, 39♂

Construct validity- hypothesis testing

Corr. VMIQ and GTVIC

r=0.72 VMIQ visual

Corr. VMIQ and MIQ

see above notes for the MIQ

Doubtful

+

See above comments for the MIQ.

Revised Version of the Vividness of Movement Imagery Questionnaire (VMIQ-2)

Sport

Roberts et al. 20081 [7]

UK

E

Athletes

351

20.44

159♀, 189♂

Construct validity- structural validity

The three-factor CTCU analysis provided the best fit to the data: χ2=840.65, df=555, CFI=0.98, NNFI=0.97, SRMR=0.04, RMSEA=0.04. Factor loadings ranged from 0.60 to 0.78. Corr. between the factors: internal and external r=0.39, internal and kinaesthetic r=0.63, external and kinaesthetic r=0.41

Very good

+

Roberts et al. reported in their article the results of three separate studies.

20081= study 1

Very good sample size for this analysis.

Sport

Roberts et al. 20082 [7]

UK

E

Athletes

355

20.44

119♀, 235♂, 1 NR

Construct validity- structural validity

The three-factor CTCU further provided the best fit to the data: χ2=1242.76, df=555, CFI=0.98, NNFI=0.97, SRMR=0.06, RMSEA=0.06. Factor loadings ranged from 0.64 to 0.82. Corr. between the factors: internal and external r=0.51, internal and kinaesthetic r=0.62, external and kinaesthetic r=0.43

Very good

+

Roberts et al. 20082 [7]= study 2

Very good sample size for this analysis.

Sport

Roberts et al. 20083 [7]

UK

E

Athletes

71

21.72

55♀, 16♂,

Construct validity- hypothesis testing

Corr. internal VMIQ-2 and visual MIQ-R

r=−0.34, p<0.05

Corr. external VMIQ-2 and visual MIQ-R

r=−0.65, p<0.01

Corr. kinaesthetic VMIQ-2 and kinaesthetic MIQ-R

r=−0.74, p<0.01

Doubtful

+

Roberts et al. 20083 [7]= study 3

Strong corr. with instruments measuring the same construct. 75% of the results are in accordance with the hypotheses.

Sport

Qwagzeh et al. 2018 [88]

JO

AR

Students

46

NR

18♀, 28♂,

Construct validity- hypothesis testing

Concurrent validity was 0.89.

Inadequate

−

No information about comparator or how concurrent validity was calculated.

Only briefly mention in the text.

Construct validity- hypothesis testing/

Known-groups validity

There were gender differences: female demonstrated more clear and vivid external imagery (p<0.001) and kinaesthetic (p<0.001) than male. For internal imagery no sign. differences (p=0.339) were found.

Inadequate

?

No adequate description provided of important characteristics of the subgroups for understanding of these results. No difference was expected.

n.d.s.

Dahm et al. 2019 [89]

AT

G

Students

254

24.0

79♀, 175♂

Construct validity- structural validity

MTMM and MT approach to CFA and three models were tested. The three-factor MTMM model provided the best fit to the data: χ2/df=1.63, CFI=0.92, SRMR=0.06, RMSEA=0.05.

Factor loadings for external 0.57–0.75, for internal 0.56–0.73, for kinaesthetic 0.60–0.74.

Very good

+

Accepted model fit: CFI or TLI >0.95, or SRMR <0.08, or RMSEA <0.06.

Not all criteria met for positive rating of this measurement property.

Wheelchair Imagery Ability Questionnaire (WIAQ)

Med

Faull & Jones 20182 [90]

UK

E

Athletes

115

31.46

62♀, 53♂

Construct validity- structural validity

CFA using maximum likelihood was performed. The three-factor 15-item model was tested using the three Bayesian Structural Equation Modelling. The interfactor correlations between the three imagery factors were as follows; external with internal r=0.71 (0.59, 0.80), external with kinaesthetic r=0.48 (0.30, 0.63), and internal with kinaesthetic r=0.63 (0.49, 0.74).

Doubtful

?

Sample size was adequate.

20172= study 2.

The use of BSEM analysis is becoming accepted as an innovative method to analyse a structural validity. However, this method was not proposed by COSMIN and therefore our rating is doubtful and indeterminate for this measures.

Med

Faull & Jones 20183 [90]

UK

E

Athletes

115

31.46

62♀, 53♂

Construct validity- hypothesis testing

Corr. WIAQ with SIAQ (total score)

external and SIAQ r=0.39

internal and SIAQ r= 0.26

kinaesthetic and SIAQ r=0.20

Corr. WIAQ and TOPS-2 (two scales, practice and competition)

external and practice r=0.23, external and competition r=0.27

kinaesthetic and practice r=0.21, kinaesthetic and competition r=0.27

No sig. corr. between internal and TOPS-2

Doubtful

+

20173= study 3.

No information about measurement properties of the comparator instrument. 75% of the results are in accordance with the hypotheses.

  1. Legend: The superscript numbers were used to distinguish the results per group
  2. Disciplines in which field the tool was evaluated: Edu education, Med medicine, Psy psychology, n.d.s. not discipline-specific; healthy participants/students
  3. Country abbreviations: AT Austria, CA Canada, CH Switzerland, ES Spain, FR France, JO Jordan, IR Iran, JP Japan, SI Slovenia, TR Turkey, NZ New Zealand, PL Poland, UK United Kingdom, USA United States of America
  4. Language of the tool: E English, F French, G German, P Portuguese, J Japanese, PO Polish, SL Slovenian, Tu Turkish, S Spanish, AR Arabic
  5. AGFI adjusted goodness of fit index, BSEM Bayesian Structural Equation Modeling, CI confidence interval, corr. correlation, CT correlated trial, CFA confirmatory factor analysis, CFI Comparative fit index, CTCU correlated trial-correlated uniqueness, COSMIN Consensus-based Standards for the selection of health Measurement Instruments Risk of Bias Checklist, df degrees of freedom, EFA exploratory factor analysis, external external perspective subscale, FOLO Functions of Observational Learning Questionnaire, FPIQ Florida Praxis Imagery Questionnaire, GFI goodness of fit index, GTVIC Gordon Test of Visual Imagery Control, internal internal perspective subscale, kinaesthetic kinaesthetic subscale, KVIQ-20 original Kinaesthetic and Visual Imagery Questionnaire, KVIQ-10 short version of the KVIQ, MS multiple sclerosis, LISREL Linear Structural RELations, MT Multi-Trait, MRT Mental Rotation Test, MTMM Multitrait-multimethod, N sample size, NR not reported, NNFI non-normed fit index, PCA principal component analysis, PD Parkinson disease, RFI Relative Fit Index, RMR the root mean square residual, RMSEA root mean square error of approximation, sign. significant, SIAQ Sport Imagery Ability Questionnaire, SRMR standardised root mean square residual, TLI Tucker-Lewis index, TAMI Test of Ability in Movement Imagery, TAMI-H Test of Ability in Movement Imagery with Hands, TAMIw* TAMI-weighted - new scoring method (More difficult questions were more weighted than relatively easier questions), TOPS-2 Test of Performance Strategies-2, visual visual subscale, VVIQ Vividness of Visual Imagery Questionnaire, χ2 chi-square;
  6. # methods could be doubtful, students received a course credits for participation. It could be interpreted that there was a certain dependency/necessity to participate, but it was not taken into account by the COSMIN evaluation
  7. Quality Criteria: ‘+’ =sufficient, ‘−’=insufficient, ‘?’=indeterminate. *See Table 1 Legend for explanation of quality criteria
  8. For criteria of an exploratory factor analysis (EFA) see de Vet et al. 2011 [52], Izquierdo et al. 2014 [61] and Watkins 2018 [62]