This loss function reflects the similarity between human choices and RL model predictions. If a human selects the left option and the RL model predicts a high probability for the left option, then \(logP_{L}\) approaches 0, causing the first term to approach 0.
Since the human chose the left option, \(B_{R}\) becomes 0, making the second term naturally zero. Therefore, the more consistent the RL model's prediction is with human choice, the closer this LL value is to 0. Conversely, it approaches negative infinity. $$ LL = \sum B_{L} \times \log P_{L} + \sum B_{R} \times \log P_{R} $$
Usage
func_logl(
i,
L_freq,
R_freq,
L_pick,
R_pick,
L_value,
R_value,
L_dir,
R_dir,
L_prob,
R_prob,
var1 = NA,
var2 = NA,
LR,
try,
value,
utility,
reward,
occurrence,
alpha,
beta
)Arguments
- i
The current row number.
- L_freq
The frequency of left option appearance
- R_freq
The frequency of right option appearance
- L_pick
The number of times left option was picked
- R_pick
The number of times left option was picked
- L_value
The value of the left option
- R_value
The value of the right option
- L_dir
Whether the participant chose the left option.
- R_dir
Whether the participant chose the right option.
- L_prob
The probability that the model assigns to choosing the left option.
- R_prob
The probability that the model assigns to choosing the left option.
- var1
[character] Column name of extra variable 1. If your model uses more than just reward and expected value, and you need other information, such as whether the choice frame is Gain or Loss, then you can input the 'Frame' column as var1 into the model.
default: var1 = "Extra_Var1"- var2
[character] Column name of extra variable 2. If one additional variable, var1, does not meet your needs, you can add another additional variable, var2, into your model.
default: var2 = "Extra_Var2"- LR
Are you calculating the probability for the left option or the right option?
- try
If the choice was random, the value is 1; If the choice was based on value, the value is 0.
- value
The expected value of the stimulus in the subject's mind at this point in time.
- utility
The subjective value that the subject assigns to the objective reward.
- reward
The objective reward received by the subject after selecting a stimulus.
- occurrence
The number of times the same stimulus has been chosen.
- alpha
[vector] Extra parameters that may be used in functions.
- beta
[vector] Extra parameters that may be used in functions.
Note
When customizing these functions, please ensure that you do not modify
the arguments. Instead, only modify the if-else statements or
the internal logic to adapt the function to your needs.
Examples
if (FALSE) { # \dontrun{
func_logl <- function(
# Trial number
i,
# Number of times this option has appeared
L_freq,
R_freq,
# Number of times this option has been chosen
L_pick,
R_pick,
# Current value of this option
L_value,
R_value,
#
L_dir,
R_dir,
#
L_prob,
R_prob,
# Extra variables
var1 = NA,
var2 = NA,
# Whether calculating probability for left or right choice
LR,
# Is it a random choosing trial?
try,
# Extra parameters
alpha,
beta
){
logl <- switch(
EXPR = LR,
"L" = L_dir * log(L_prob),
"R" = R_dir * log(R_prob)
)
}
} # }