Table 2 Interrelations between dopamine synthesis and levels of GABA and glutamate.

	Group (AN-FEP vs HC)		GABA levels in ACC		Group*Neuro-transmitter^a		Glutamate levels in thalamus		Glutamate levels in ACC		Adjusted R-squared (%)
	Coef.	p-value	Coef.	p-value	Coef.	p-value	Coef.	p-value	Coef.	p-value
Model 1* k₃ = Group + GABA + Group*GABA + sex	−0.0052	0.78	0.008	0.67	0.06	0.002	–	–	–	–	14.1%
Model 2* k₃ = Group + GABA + Group*GABA + Glu thal + sex	−0.0011	0.96	0.01	0.64	0.06	0.007	0.0323	0.11	–	–	17.4%
Model 3 k₃ = Group + Group*GABA + Glu ACC + sex	−0.0069	0.73	0.01	0.63	0.06	0.003	–	–	−0.0162	0.39	13.3%
Model 4 k₃ = Group + Group*Glu thal + Glu thal + sex	−0.0045	0.82	–	–	−0.003	0.90	0.0417	0.04	–	–	0.8%
Model 5 k₃ = Group + Group*Glu ACC + Glu ACC + sex	−0.0081	0.68	–	–	0.03	0.16	–	–	−0.0124	0.53	2.1%

Multiple linear regression models of the relation between dopamine synthesis in Nucleus Accumbens (k₃) and different combinations of GABA levels in anterior cingulate cortex (ACC), glutamate levels in ACC, and glutamate levels in thalamus in antipsychotic-naïve first-episode patients with psychosis (AN-FEP) and healthy controls (HC). The dependent and independent variables are outlined for the separate models (without the intercept and β-coefficients due to space limitations). Sex was included as a covariate but did not contribute to any of the models.
AN-FEP antipsychotic-naïve first-episode patients with psychosis, HC healthy controls, ACC anterior cingulate cortex, Glu glutamate, thal thalamus.
*For model 1 and 2 indicates that the overall models were statistically significant. Significant p-values for the independent variables are highlighted with bold.
*p < 0.05.
^aGroup*GABA for model 1, 2 and 4; group*glu thal for model 3; and group*glu ACC for model 5.

Search