Hostname: page-component-7b7567c8c-7pwfn Total loading time: 0 Render date: 2026-11-16T18:07:11.020Z Has data issue: false hasContentIssue false

INABILITY AND OBLIGATION IN INTELLECTUAL EVALUATION

Published online by Cambridge University Press:  21 December 2018

Rights & Permissions [Opens in a new window]

Abstract

If moral responsibilities prescribe how agents ought to behave, are there also intellectual responsibilities prescribing what agents ought to believe? Many theorists have argued that there cannot be intellectual responsibilities because they would require the ability to control whether one believes, whereas it is impossible to control whether one believes. This argument appeals to an “ought implies can” principle for intellectual responsibilities. The present paper tests for the presence of intellectual responsibilities in social cognition. Four experiments show that intellectual responsibilities are attributed to believe things and that these responsibilities can exceed what agents are able to believe. Furthermore, the results show that agents are sometimes considered responsible for failing to form true beliefs on the basis of good evidence, and that this effect does not depend on the seriousness of the consequences for failing to form a belief. These findings clarify when and how responsibilities for belief are attributed, falsify a conceptual entailment between ability and responsibility in the intellectual domain, and emphasize the importance of objective truth in intellectual evaluations.

Information

Type
Articles
Copyright
Copyright © Cambridge University Press 2018
Figure 0

Fig. 1.Experiment 1. Mean percentage choosing each OIC test item option grouped by ability.

Figure 1

Table 1Experiment 1. Multivariate analysis of variance, Scenario, Wilks' Lambda = 0.870, F(3, 159) = 7.95, p < 0.001, ηp2 = 0.130, Ability, Wilks' Lambda = 0.344, F(3, 159) = 100.88, p < 0.001, ηp2 = .656, S*A, Wilks' Lambda = 0.988, F(3, 159) = 0.64, p = 0.589, ηp2 = 0.012.

Figure 2

Table 2Experiment 1. Independent samples t-tests by ability.

Figure 3

Table 3Experiment 1. Independent samples t-tests by scenario.

Figure 4

Table 4Experiment 1. One sample t-tests. Test value = 4.

Figure 5

Table 5Experiment 2. Multivariate analysis of variance, Scenario, Wilks' Lambda = 0.914, F(4, 162) = 3.82, p = 0.005, ηp2 = 0.086, Ability, Wilks' Lambda = 0.322, F(4, 162) = 85.12, p <0.001, ηp2 = 0.678, S*A, Wilks' Lambda = 0.987, F(4, 162) = .53, p = 0.715, ηp2 = 0.013.

Figure 6

Table 6Experiment 2. Multivariate analysis of variance controlling for blame, Scenario, Wilks' Lambda = 0.960, F(3, 162) = 2.27, p = 0.083, ηp2 = 0.040, Ability, Wilks' Lambda = 0.344, F(3, 162) = 103.20, p <0.001, ηp2 = 0.656, S*A, Wilks' Lambda = 0.987, F(3, 162) = 0.703, p = 0.551, ηp2 = 0.013, Blame, Wilks' Lambda = 0.666, F(3, 162) = 27.08, p <0.001, ηp2 = 0.334.

Figure 7

Fig. 2.Experiment 2. Panel A: mean response to test statements about whether an agent had an intellectual responsibility, deserves blame, and had a possibility to believe. Scales ran 1 (SD)–7 (SA). Panel B: mean estimate of the percentage chance that the agent would form the belief. Scales ran 0–100%. All error bars represent ± 1 SEM.

Figure 8

Table 7Experiment 2. Independent samples t-tests by ability.

Figure 9

Table 8Experiment 2. One sample t-tests. Test value = 4.

Figure 10

Table 9Experiment 3. Logistic regression predicting responses to the OIC test item (2-option). The full model was statistically significant, χ2 (3, N = 162) = 14.64, p <0.01, and explained between 9% and 12% of the variance in response to the test item.

Figure 11

Table 10Experiment 3. Multivariate analysis of variance, Evidence, Wilks' Lambda = 0.537, F(3, 156) = 44.83, p <0.001, ηp2 = 0.463, Truth, Wilks' Lambda = 0.872, F(3, 156) = 7.65, p <0.001, ηp2 = 0.128, E*T, Wilks' Lambda = 0.997, F(3, 156) = 0.133, p = 0.940, ηp2 = 0.003.

Figure 12

Table 11Experiment 3. Independent samples t-tests by evidence.

Figure 13

Table 12Experiment 3. Independent samples t-tests by truth-value.

Figure 14

Table 13Experiment 3. One sample t-tests. Test value = 4.

Figure 15

Table 14Experiment 4. Logistic regression predicting responses to the OIC test item (2-option). The full model was statistically significant, χ2 (3, N = 185) = 50.63, p < 0.001, and explained between 24% and 32% variance in response to the test question.

Figure 16

Table 15Experiment 4. Multivariate analysis of variance, Stakes, Wilks' Lambda = 0.640, F(3, 179) = 33.62, p <0.001, ηp2 = 0.360, Epistemic, Wilks' Lambda = 0.721, F(3, 179) = 23.13, p <0.001, ηp2 = 0.279, S*E, Wilks' Lambda = 0.976, F(3, 179) = 1.47, p = 0.225, ηp2 = 0.024.

Figure 17

Table 16Experiment 4. Evidence independent samples t-tests by epistemic status.

Figure 18

Table 17Experiment 4. Independent samples t-tests by stakes.

Figure 19

Table 18Experiment 4. One sample t-tests. Test value = 4.