LEARNrNQ BY MARINE INVERTEBRATES
17
tinguishable from normal octopuses. Removal of the superior frontal
lobe has a similar effect. It is assumed that the residual capacity to
learn is at least partly due to structures in the optic lobes, which
cannot be removed without blinding the animals.
Similarly, it has been found that removal of the inferior frontal and
subfrontal lobes abolishes touch learning whilst having no other
detectable effect on the animals’ behaviour. Partial lesions to the
vertical and subfrontal lobes have effects proportional to the amount
taken out, with, in the case of the subfrontal lobe, some indication that
a minimum number of cells is required for satisfactory performance in
any particular task. Approximately half a million subfrontal cells must
be left if the animal is to discriminate effectively between a rough and a
smooth perspex cylinder, while a few thousand suffice for learning to
reject either when it is repeatedly presented (Wells, 1959). At the present
ti.me it appears that, while removal of the vertical and/or superior
frontal lobes measurably slows touch learning, removal of the inferior
frontal and subfrontal lobes is without any such effect on visual
responses.
Reviews of the results of brain lesion experiments up to 1960-61 are
included in Wells (19624 and Young (1961).
Most recent work has been concentrated in an attempt to define
more closely the function of the vertical and superior frontal lobes.
Thus Maldonado (1963 a, b, c ; 1964) has made a careful analysis of the
effect of vertical lobe removal on learning to attack crabs and other
objects seen at a distance. By accurate automatic measurement of the
times taken to attack and the nature of the approach to the prey before
and after the operation he showed (1) that both the time taken to
complete an attack and the variability of this time from trial to trial
increased as an immediate result of the operation, and (2) that after the
operation details of the performance-the nature of the approach and
the time delays-reverted to the condition seen in the same animals at
the start of training ; performance improved with further practice but
they never became so swift and accurate as before. The effects were
most marked in the animals with the largest lesions.
Maldonado’s results show that the vertical lobe plays some part in
positive learning as well as in learning not to respond. In this they
confirm the results of a number of previous experiments in which
vertical lobe removal has been followed by a drop in the number and
speed of attacks on figures other than crabs which have hitherto
appeared to constitute a special case (see 11, A, 6, p. 12). It seems in
general that vertical lobe removal reduces the probability of previous
experience affecting responses. Individuals that attacked a lot before the
17
tinguishable from normal octopuses. Removal of the superior frontal
lobe has a similar effect. It is assumed that the residual capacity to
learn is at least partly due to structures in the optic lobes, which
cannot be removed without blinding the animals.
Similarly, it has been found that removal of the inferior frontal and
subfrontal lobes abolishes touch learning whilst having no other
detectable effect on the animals’ behaviour. Partial lesions to the
vertical and subfrontal lobes have effects proportional to the amount
taken out, with, in the case of the subfrontal lobe, some indication that
a minimum number of cells is required for satisfactory performance in
any particular task. Approximately half a million subfrontal cells must
be left if the animal is to discriminate effectively between a rough and a
smooth perspex cylinder, while a few thousand suffice for learning to
reject either when it is repeatedly presented (Wells, 1959). At the present
ti.me it appears that, while removal of the vertical and/or superior
frontal lobes measurably slows touch learning, removal of the inferior
frontal and subfrontal lobes is without any such effect on visual
responses.
Reviews of the results of brain lesion experiments up to 1960-61 are
included in Wells (19624 and Young (1961).
Most recent work has been concentrated in an attempt to define
more closely the function of the vertical and superior frontal lobes.
Thus Maldonado (1963 a, b, c ; 1964) has made a careful analysis of the
effect of vertical lobe removal on learning to attack crabs and other
objects seen at a distance. By accurate automatic measurement of the
times taken to attack and the nature of the approach to the prey before
and after the operation he showed (1) that both the time taken to
complete an attack and the variability of this time from trial to trial
increased as an immediate result of the operation, and (2) that after the
operation details of the performance-the nature of the approach and
the time delays-reverted to the condition seen in the same animals at
the start of training ; performance improved with further practice but
they never became so swift and accurate as before. The effects were
most marked in the animals with the largest lesions.
Maldonado’s results show that the vertical lobe plays some part in
positive learning as well as in learning not to respond. In this they
confirm the results of a number of previous experiments in which
vertical lobe removal has been followed by a drop in the number and
speed of attacks on figures other than crabs which have hitherto
appeared to constitute a special case (see 11, A, 6, p. 12). It seems in
general that vertical lobe removal reduces the probability of previous
experience affecting responses. Individuals that attacked a lot before the