An associative network is constructed with chaotic neuron models inter
connected through a conventional auto-associative matrix of synaptic w
eights. The associative dynamics of the network is analysed with spati
otemporal output patterns, quasi-energy function, distances between in
ternal state vectors and orbital instability. The network shows a peri
odic response after a nonperiodic transient phase when the external st
imulations are spatially constant. The retrieval characteristics and t
he duration in the transient phase are dependent on the initial condit
ions. The results imply that the transient dynamics can be interpreted
as a memory searching process. The network also shows periodic respon
ses with short and very long periods when external stimulations are no
t spatially constant, but corresponding to a stored pattern and an uns
tored pattern, respectively. The responses to the external stimulation
s can be utilized for a pattern recognition with nonlinear dynamics. C
opyright (C) 1996 Elsevier Science Ltd.