Table 2.

Heuristic choice models (Wulff and van den Bos, 2018): The ITCH model contains four free parameters: βxA is the relevancy of absolute reward, βxR is the relevancy of relative reward, βtA is the relevancy of absolute delay, and βtR is the relevancy of relative delay

ITCHDRIFTTRADE
Factor 1βxA*(x2−x1) βxA*z(x2−x1) v(x)=(1scaling2)*log(1+scaling2*x)
Factor 2βxR*(x2−x1)(x2+x1)2 βxR*z(x2−x1x1) (v(x2)−v(x1))
Factor 3βtA*(t2−t1) βtA*z(t2−t1) w(x)=(1scaling3)*log(1+scaling3*x)
Factor 4βtR*(t2−t1)(t2+t1)2 βxt*z((x2x1)1t2−t1−1) −scaling1*(w(t2)−w(t1))
SV(DEL)-SV(Immediate) = Factor 1 + Factor 2 +  Factor 3 + Factor 4Factor 1 + Factor 2 +  Factor 3 + Factor 4Factor 2 − Factor 4

  • For the DRIFT model, βxA and βtA are equivalent to the ITCH model, but βxR is additionally scaled by the amount of immediate reward. βxt is a proportional distance factor that increases with larger differences in reward amount and decreases with larger differences in delay. TRADE utilizes a logarithmic scaling function for reward amount and delay (see Factors 1 and 3), where Scaling 2 scales both reward amounts and Scaling 3 scales absolute and relative delays. Finally, Scaling 1 scales the relative delays for their relevance.