Table 3.

Glossary of key terms describing the full model in Fig. 8 and in Eqs. 1–12

$V_e(t)$	Learned value function (expectation) in [0,1] at time step t where $e\mathit{ϵ}\{m,o\}$ and m stands for magnitude valuation of the stimulus, o stands for omission probability valuation of the stimulus. This is visualized in the nodes in the Critic in Fig. 8
${\mathit{\theta }}_e(t)$	Gives the parameter (Critic weights) indexed by e that valuate the stimuli at each time step t. These weights are denoted by (1) on Fig. 8
${\mathit{\delta }}_m(t)$	Is the prediction error generated by Critic that updates the ${\mathit{\theta }}_m(t)$ parameters of the magnitude Critic
${\mathit{\delta }}_o(t)$	Is the prediction error generated by Critic that updates the ${\mathit{\theta }}_o(t)$ parameters of the omission Critic
$u_r(t)$	Stimulus/response option neural-dynamic variables indexed by $r\mathit{ϵ}[1,R]$ where $R\mathit{ϵ}\{S1,S2,R1,R2\}$
Rew(t)	Reward expectation (right-side blue node Fig. 8) classification of stimulus
Om(t)	Omission expectation (left-side blue node Fig. 8) classification of stimulus
$x_s(t)$	Meta-parameter that controls the slope and threshold of Rew and Om variables allowing for competition for stimulus classification
${\mathit{\Omega }}_{kl}$	Connection strength in [0, 1] between pre-synaptic node $k\mathit{ϵ}\{Om,Rew\}$ and post-synaptic node $l\mathit{ϵ}\{R1,R2\}$. See connections denoted by (2) on Fig. 8