Selection for Reinforcement-Free Learning Ability as an Organizing Factor in the Evolution of Cognition

Figure 11

Attempted idealization of the solution evolved in runs 4 and 9 of the main model. These patterns were named TF (for turn or Forward), LR (for left or Right) and SJ (for step or Jump). TF promotes the step and jump actions when there is vertically oriented prey straight ahead and both turn actions when there is horizontally oriented prey to a side. SJ works as in Figure 10, assessing whether it is better to step or jump, such that the correct forward action is picked in case TF selects forward movement. LR promotes leftward turns and inhibits right turns when there is horizontally oriented prey to the left and not to the right and vice versa, such that the correct turn direction is picked in case TF selects to make a turn. Blue connections may vary.