Canonical Correlation Analysis (CCA) describes a family of methods useful in identifying the links between data from different modalities. CCA simultaneously evaluates two different sets of variables, identifying the sources of common variation across the paired high-dimensional datasets. We apply CCA to jointly model natural language reports of psychedelic experiences paired with receptor affinity for 27 different hallucinogenic molecules. Psychedelic drugs are being embraced by researchers as treatments for mental health conditions, but the mechanism of action of these drugs remains the subject of intense inquiry. The quality of the acute drug experience appears to predict the long-term efficacy of these treatments. This suggests better characterizations of the psychedelic experience may inform their therapeutic use. Towards that end, we use CCA to reveal the common structure underlying each drug's unique receptor affinity fingerprint with its phenomenological flavor as captured in subjective testimonials.