It is important to be able to predict a response to lithium before lithium administration because it usually takes 2 or 3 weeks for lithium to manifest its effect sufficiently. We hypothesized that lithium responders have a high post-synaptic serotonergic receptor function, whereas non-responders have low one. This is because it has been suggested that lithium's primary actions on the serotonergic neuron may be presynaptic and several reports have shown that lithium enhances serotonin synthesis and secretion at the presynaptic serotonergic neuron. As a preliminary study, we investigated the relationship between several responses to meta-chlorophenylpiperazine (a serotonergic agonist) reflecting the serotonergic receptor function and the clinical lithium response of 10 patients. No significant relationship between hormone responses and improvement ratios was found. However, there were significant relationships found between 'anger' during the challenge test and the improvement ratio and between 'tension' during the test and the improvement ratio. Thus, it seems that the hormone response to a serotonergic agonist is not useful for predicting lithium response. However, subjective responses such as anger and tension to a serotonergic agonist are possible candidates for predicting lithium response. In any case, further studies with larger numbers of participants are required to investigate whether subjective responses to a serotonergic agonist are useful for predicting lithium response.