Using behavior-analytic implicit tests to assess sexual interests among normal and sex-offender populations

Figures & data

Fig. 1.  Total correct responses during consistent and inconsistent IAT test blocks for all groups of subjects. The difference in response accuracy across the two blocks is indicative of the size of the IAT effect. Note that the differential is in the same direction (consistent greater than inconsistent) for all male subject groups but not for the female subject group.

Table 1. Total number of correct responses on both blocks of a modifed IAT for a sample of offenders and control subjects. C refers to consistent test blocks and I refers to inconsistent test blocks. A positive difference score is indicative of a pre-exisitng child-sex association.

P	Female control		Male control		Male offender (non-sexual)		Offender against adult		Internet offender		Offender against child
	C	I	C	I	C	I	C	I	C	I	C	I
Total correct responses
1	48	66	75	78	48	66	47	29	40	33	55	56
2	75	78	56	57	59	57	33	38	49	64	60	50
3	62	69	74	78	54	48	59	48	43	37	62	53
4	77	76	67	45	52	52	30	36	64	48	43	40
5	65	65	68	51	57	64	40	54	72	74	30	16
6	18	37	55	58	58	58	55	56	69	40	64	60
7	57	55	62	59	61	58	47	49	27	38	54	51
8	53	50	53	48	54	30	54	53	59	58	58	50
9	57	57	54	50	59	50	74	47	56	47	37	38
10	59	55	26	50	65	61	30	32	58	69	61	57
Mean total correct responses
	57.1	60.8	59.0	57.4	56.7	54.4	46.9	44.2	53.7	50.8	52.4	47.1
Difference score
	−3.7	+1.6	+2.3	+2.7	+2.9	+5.3