Table 3 OLS regressions with rejection rates as the dependent variable.

From: GPT-3.5 altruistic advice is sensitive to reciprocal concerns but not to strategic risk

 

All

Temp = 0.5

Temp = 1.0

Temp = 1.5

Fairness

0.039***

(0.006)

0.000

(0.0001)

0.010*

(0.005)

0.107***

(0.018)

UG—Receiver

4.904***

(0.090)

0.214***

(0.031)

3.954***

(0.139)

10.560***

(0.224)

Fairness × UG—Receiver

− 1.010***

(0.098)

− 0.078*

(0.039)

− 0.988***

(0.154)

− 1.980***

(0.245)

Prompted demographics

− 0.090

(0.057)

0.000

(0.021)

− 0.097

(0.087)

− 0.175

(0.142)

Age (31–50)

− 0.313***

(0.055)

− 0.027

(0.019)

− 0.307***

(0.083)

− 0.610***

(0.139)

Age (51–70)

0.209***

(0.051)

0.010

(0.016)

0.217***

(0.076)

0.399***

(0.131)

Male

0.661***

(0.056)

0.063**

(0.021)

0.671***

(0.085)

1.252***

(0.141)

Non-binary

− 0.954***

(0.031)

− 0.043***

(0.010)

− 0.816***

(0.046)

− 2.005***

(0.077)

Constant

0.585***

(0.082)

0.056

(0.032)

0.713***

(0.127)

0.993***

(0.205)

Observations

299,578

100,000

99,949

99,629

R2

0.022

0.001

0.019

0.047

Adjusted R2

0.022

0.001

0.019

0.047

  1. Note. Rejection likelihood increases significantly with unfair offers and varies with temperature settings. UG – Receiver = dummy variable capturing suggestions to the receiver in the Ultimatum Game, Fairness = amount of money to the receiver; *p < 0.05; **p < 0.01; ***p < 0.001.