In order to explore the influences of nonlinear phenomena (NLP) on the vocal responses in concave-eared torrent frogs (Odorrana tormota), and whether NLP can enhance the unpredictability of calls, strange male concave-eared torrent frogs were stimulated by replaying NLP and linear calls (Fig. 1). The numbers of evoked vocal responses were recorded and the spectral and temporal parameters of response calls were analyzed. The results revealed that NLP calls could cause more response calls in strange frogs (n = 22, P < 0.05, Table 2), but there was no significant difference in the first response latency (P > 0.05, Table 3) of two kinds of evoked response calls. We then employed Praat to divide the calls into different temporal segments and obtain various temporal and spectral parameters. Wilcoxon’s signed rank test was used to determine whether there was significant difference in sound parameters of two kinds of evoked response calls. The results showed that the total duration of response calls stimulated by linear calls was longer than that of NLP-stimulated ones, while other vocal parameters, such as average fundamental frequency, maximum fundamental frequency, minimum fundamental frequency and dominant frequency, showed no significant difference (P > 0.05, Table 4). It was possible that strange male frogs would respond to the conspecific calls instantly in order to protect their territory and resources, however, there were no differences in response time to NLP calls and linear calls. In terms of response call numbers, more responses were stimulated by NLP calls which contain more complicated information, and thus increase the attention of the receiver. The current study reveals that the NLP calls will enhance vocal unpredictability and cause more responses of strange male frogs.