The Effect of Stimulus-Response Compatibility on the Association of Fluid Intelligence and Working Memory with Choice Reaction Times

Gizem Hülür; Doris Keye-Ehing; Klaus Oberauer; Oliver Wilhelm

doi:10.5334/joc.66

The Effect of Stimulus-Response Compatibility on the Association of Fluid Intelligence and Working Memory with Choice Reaction Times

Journal of Cognition

Volume 2 (2019): Issue 1

By: Gizem Hülür , Doris Keye-Ehing, Klaus Oberauer and Oliver Wilhelm

Open Access

|Jun 2019

Figures & Tables

Stimulus-response sets for four-choice reaction time tasks in Wilhelm & Oberauer (2006).

Stimulus-response sets for two-choice reaction time tasks in Study 1.

Table 1

Descriptive Statistics for Latencies and Accuracies in the Choice Reaction Time Tasks in Study 1.

	Latencies						Accuracies
	Mean	SD	Min	Max	Skewness	Kurtosis	Mean	SD	Min	Max	Skewness	Kurtosis
Arrow task
Compatible, speed instruction	314	35	239	423	0.12	–0.02	.82	0.10	.50	1.00	–0.77	0.20
Compatible, accuracy instruction	360	33	287	474	0.66	0.70	.96	0.04	.81	1.00	–1.46	2.28
Arbitrary, speed instruction	312	44	208	441	0.19	0.24	.77	0.11	.50	1.00	–0.49	–0.46
Arbitrary, accuracy instruction	412	61	296	594	0.80	0.14	.96	0.04	.78	1.00	–1.77	3.29
Shape task										1.00
Compatible, speed instruction	240	20	186	312	0.34	0.70	.86	0.09	.62	1.00	–0.53	–0.42
Compatible, accuracy instruction	286	39	217	460	1.49	3.21	.99	0.02	.92	1.00	–1.88	3.40
Arbitrary, speed instruction	326	43	228	435	0.05	–0.16	.80	0.12	.51	1.00	–0.55	–0.61
Arbitrary, accuracy instruction	385	45	300	607	1.41	4.14	.96	0.03	.85	1.00	–1.14	1.50
Word task
Compatible, speed instruction	333	42	225	475	0.04	1.57	.79	0.10	.52	.99	–0.41	–0.44
Compatible, accuracy instruction	413	47	329	580	1.10	1.59	.96	0.04	.81	1.00	–1.46	2.40
Arbitrary, speed instruction	330	48	225	548	0.62	2.23	.76	0.11	.50	.96	–0.32	–0.56
Arbitrary, accuracy instruction	414	44	337	579	0.92	1.01	.96	0.04	.79	1.00	–1.59	3.53

[i] Note: N = 135, M = mean, SD = standard deviation, Min = Minimum, Max = Maximum, Latencies in milliseconds, accuracies in proportion of correct responses.

Table 2

Experimental Effects on Latencies and Accuracies in Study 1: Results from Repeated Measure ANOVAs.

	Latencies					Accuracies
	df (hypothesis)	df (error)	F	p	partial eta-squared	df (hypothesis)	df (error)	F	p	partial eta-squared
Arrow task
SRC	1	134	76.80	<.01	.36	1	134	40.84	<.01	.23
Instruction	1	134	448.64	<.01	.77	1	134	383.67	<.01	.74
SRC × Instruction	1	134	146.46	<.01	.52	1	134	29.53	<.01	.18
Shape task
SRC	1	134	1394.70	<.01	.91	1	134	73.22	<.01	.35
Instruction	1	134	291.98	<.01	.69	1	134	333.91	<.01	.71
SRC × Instruction	1	134	8.94	<.01	.06	1	134	11.73	<.01	.08
Word task
SRC	1	134	0.27	.60	.00	1	134	16.48	<.01	.11
Instruction	1	134	418.07	<.01	.76	1	134	515.17	<.01	.79
SRC × Instruction	1	134	0.88	.35	.01	1	134	14.32	<.01	.10

[i] Note: N = 135. SRC = stimulus-response compatibility.

Measurement models for reaction time tasks in Study 1 and standardized parameter estimates. (a) Model A including a general reaction time factor. Model fit: χ²[df] = 151.76[20], p < .01, CFI = .75, RMSEA = .22, SRMR = .12. (b) Model B with a nested factor reflecting the effects of the instruction (speed vs. accuracy). Model fit: χ²[df] = 47.17[16], p < .01, CFI = .94, RMSEA = .12, SRMR = .06. (c) Model C with an additional nested factor reflecting binding costs in arbitrary conditions. Model fit: χ²[df] = 19.92[12], p = .07, CFI = .99, RMSEA = .07, SRMR = .04. Models only included indicators of reaction time in the arrow and shape tasks because no consistent stimulus-response compatibility effects were found in the word task. Model B fit the data better than Model A and Model C fit the data better than Model B. N = 135, * p < .05.

Structural model for examining associations of reaction time with fluid intelligence and working memory in Study 1. Model fit: χ² = 129.59, df = 89, p < .01, CFI = 0.95, RMSEA = .06, SRMR = .06, N = 135, * p < .05.

Table 3

Descriptive Statistics and Correlations of Variables Included in the Structural Model (Study 1).

	M	SD	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)	(11)	(12)	(13)	(14)	(15)	(16)
(1) ARR/CO/SP	314	35	1
(2) SHA/CO/SP	240	20	.53*	1
(3) ARR/AR/SP	312	44	.62*	.50*	1
(4) SHA/AR/SP	326	43	.51*	.36*	.67*	1
(5) ARR/CO/AC	360	33	.48*	.36*	.20*	.28*	1
(6) SHA/CO/AC	286	39	.41*	.38*	.27*	.30*	.68*	1
(7) ARR/AR/AC	412	61	.44*	.36*	.42*	.47*	.62*	.64*	1
(8) SHA/AR/AC	385	45	.35*	.31*	.29*	.36*	.59*	.63*	.62*	1
(9) Arrow Series	61	14	–.09	–.14	–.11	–.12	.02	.05	–.07	–.06	1
(10) Number Series	84	17	.01	–.20*	.02	–.15	–.09	–.18*	–.20*	–.25*	.16	1
(11) Propositions	40	15	–.07	–.07	–.19*	–.14	–.01	–.03	–.07	–.05	.07	.21*	1
(12) Raven’s Matrices	57	18	–.11	–.19*	–.08	–.21*	–.01	.03	–.15	–.09	.32*	.31*	.37*	1
(13) BIS	50	13	–.25*	–.20*	–.14	–.23*	–.19*	–.10	–.23*	–.22*	.30*	.47*	.33*	.49*	1
(14) Rotation Span	75	15	–.18*	–.22*	–.19*	–.22*	.00	.01	–.20*	–.24*	.28*	.33*	.20*	.45*	.36*	1
(15) Counting Span	85	12	–.07	–.15	–.19*	–.17*	–.01	–.04	–.10	–.11	.27*	.39*	.26*	.32*	.41*	.42*	1
(16) Memory Updating	65	11	–.19*	–.19*	–.14	–.22*	–.13	–.16	–.17*	–.28*	.26*	.38*	.15	.27*	.45*	.39*	.31*	1

[i] Note: N = 135, M = mean, SD = standard deviation. Variables were z-standardized before the analysis. ARR = arrow, SHA = shape, CO = compatible, AR = arbitrary, gf = fluid intelligence, WMC = working memory capacity. Latencies in milliseconds (Variables 1 to 8). gf and WM performance in percentage correct responses (variables 9 to 16). Correlations equal to or greater than |.17| are significantly different from 0 at p < .05.

Stimulus-response sets for (a) two-choice and (b) four-choice reaction time tasks in Study 2.

Table 4

Descriptive Statistics for Latencies and Accuracies in the Choice Reaction Time Tasks in Study 2.

	Latencies						Accuracies
	Mean	SD	Min	Max	Skewness	Kurtosis	Mean	SD	Min	Max	Skewness	Kurtosis
Arrow task
2-choice (left-right) compatible	357	29	293	555	2.01	11.82	.93	0.05	.70	1.00	–1.51	2.65
2-choice (left-right) arbitrary	399	44	282	563	0.72	1.29	.91	0.06	.63	1.00	–1.65	4.04
2-choice (up-down) compatible	365	31	303	490	0.83	1.23	.92	0.06	.68	1.00	–1.51	3.39
2-choice (up-down) arbitrary	391	40	316	555	0.90	1.57	.91	0.06	.57	.99	–1.80	5.03
4-choice compatible	431	42	355	569	0.66	0.21	.94	0.05	.53	1.00	–3.54	20.38
4-choice arbitrary	588	93	445	914	1.41	2.11	.90	0.06	.60	.99	–1.48	4.04
Word task
2-choice (left-right) compatible	402	33	334	499	0.56	0.11	.89	0.07	.58	.99	–1.44	3.43
2-choice (left-right) arbitrary	400	39	312	572	0.95	1.90	.90	0.06	.65	.99	–1.45	2.56
2-choice (up-down) compatible	422	41	303	539	0.36	0.16	.89	0.06	.60	1.00	–1.65	4.47
2-choice (up-down) arbitrary	423	43	330	599	0.75	1.15	.89	0.06	.63	.99	–1.24	1.76
4-choice compatible	565	66	452	884	1.29	3.10	.89	0.08	.41	.98	–3.18	14.54
4-choice arbitrary	645	95	497	936	1.03	0.68	.88	0.09	.42	.98	–3.13	12.75

[i] Note: N = 153, M = mean, SD = standard deviation, Min = minimum, Max = maximum, Latencies in milliseconds, accuracies in proportion of correct responses.

Table 5

Experimental Effects on Latencies and Accuracies: Results from Repeated Measure ANOVAs in Study 2.

	Latencies					Accuracies
	Df (hypothesis)	df (error)	F	p	Partial eta-squared	df (hypothesis)	df (error)	F	p	Partial eta-squared
Arrow task (left/right)
SRC	1	152	574	<.01	.79	1	152	84.48	<.01	.36
# response alternatives	1	152	1267	<.01	.89	1	152	0.35	.56	.00
SRC × # response alternatives	1	152	323	<.01	.68	1	152	16.18	<.01	.10
Arrow task (up/down)
SRC	1	152	527	<.01	.78	1	152	70.62	<.01	.32
# response alternatives	1	152	1402	<.01	.90	1	152	2.59	.11	.02
SRC × # response alternatives	1	152	430	<.01	.74	1	152	32.69	<.01	.18
Word task (left/right)
SRC	1	152	148	<.01	.49	1	152	0.07	.79	.00
# response alternatives	1	152	2146	<.01	.93	1	152	4.71	.03	.03
SRC × # response alternatives	1	152	252	<.01	.62	1	152	16.84	<.01	.10
Word task (up/down)
SRC	1	152	164	<.01	.52	1	152	5.47	.02	.03
# response alternatives	1	152	1845	<.01	.92	1	152	0.38	.54	.00
SRC × # response alternatives	1	152	221	<.01	.59	1	152	4.77	.03	.03

[i] Note: N = 153, SRC = stimulus-response compatibility.

Measurement models for reaction time tasks in Study 2. (a) Model A including a general reaction time factor. Model fit: χ²[df] = 43.14[8], p < .01, CFI = .93, RMSEA = .17, SRMR = .05. (b) Model B with a nested factor reflecting binding costs in arbitrary conditions. Model fit: χ²[df] = 3.49[5], p = .63, CFI = 1.00, RMSEA = .00, SRMR = .01). Models only included indicators of reaction time in the arrow task. Model B fit the data better than Model A. N = 153, * p < .05.

Structural model for examining associations between reaction time and intelligence in Study 2. Model fit: χ² = 72.11, df = 56, p = .07, CFI = 0.98, RMSEA = .04, SRMR = .06, N = 153, * p < .05.

Table 6

Descriptive Statistics and Intercorrelations of Variables Included in the Structural Model in Study 2.

	M	SD	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)	(11)	(12)	(13)
(1) LR/CO	357	29	1
(2) LR/AR	399	44	.55*	1
(3) UD/CO	365	31	.66*	.71*	1
(4) UD/AR	391	40	.51*	.79*	.67*	1
(5) 4C/CO	431	42	.51*	.57*	.68*	.60*	1
(6) 4C/AR	588	93	.32*	.57*	.42*	.59*	.44*	1
(7) gf: Equations	56	22	–.18*	–.25*	–.18*	–.09	–.11	–.16*	1
(8) gf: Propositions	48	22	–.18*	–.30*	–.20*	–.18*	–.22*	–.09	.17*	1
(9) gf: Matrices	54	21	–.12	–.14	–.16	–.08	–.12	–.16*	.37*	.31*	1
(10) gf: Raven’s Matrices	57	20	–.04	–.14	–.02	.00	–.04	–.01	.33*	.28*	.49*	1
(11) gc: Figural	62	13	–.17*	–.13	–.08	–.06	–.12	–.03	.25*	.14	.17*	.27*	1
(12) gc: Numeric	56	19	.06	–.07	.07	–.01	.05	.00	.21*	.07	.22*	.23*	.34*	1
(13) gc: Verbal	65	13	–.11	–.12	–.10	–.03	–.12	–.05	.18*	.16	.15	.25*	.49*	.50*	1

[i] Note: N = 153. Variables were z-standardized before the analysis. M = mean, SD = standard deviation, LR = left/right, UD = up/down, CO = compatible, AR = arbitrary, gf = fluid intelligence, gc = crystallized intelligence. Latencies in milliseconds. gf and gc performance in percentage correct responses. * p < .05.

References

Authors

Metrics

Articles in this issue

DOI: https://doi.org/10.5334/joc.66 | Journal eISSN: 2514-4820

Journal RSS Feed

Language: English

Submitted on: Nov 30, 2018

Accepted on: May 6, 2019

Published on: Jun 11, 2019

Published by: Ubiquity Press

In partnership with: Paradigm Publishing Services

Publication frequency: 1 issue per year

Keywords:

Response speed,

Working memory,

Reasoning

© 2019 Gizem Hülür, Doris Keye-Ehing, Klaus Oberauer, Oliver Wilhelm, published by Ubiquity Press
This work is licensed under the Creative Commons Attribution 4.0 License.

Volume 2 (2019): Issue 1

The Effect of Stimulus-Response Compatibility on the Association of Fluid Intelligence and Working Memory with Choice Reaction Times

Figures & Tables

Figure 1

Figure 2

Table 1

Table 2

Figure 3

Figure 4

Table 3

Figure 5

Table 4

Table 5

Figure 6

Figure 7

Table 6

Paradigm

My account