The study explores how dopamine (DA) neurons in the medial and lateral ventral tegmental area (VTA) encode an animal's behavioral state and its rate of change during a reward-seeking task. It reveals that two distinct DA neuron subtypes encode either the current state or its rate of change. Medial VTA (mVTA) DA neurons show sustained activity, while lateral VTA (lVTA) DA neurons exhibit transient activity. These activity patterns align with DA release patterns in anatomically defined mesoaccumbal pathways. The findings suggest that the brain uses two parallel lines for proportional-differential encoding of a state variable and its temporal dynamics, similar to sensory systems. The study also shows that DA cell body activity directly translates to DA release within anatomically defined mesoaccumbal subsystems. Separate DA pathways encode reward expectation and reward prediction error (RPE). The medial nucleus accumbens (NAcMed) primarily encodes reward expectation, while the lateral nucleus accumbens (NAcLat) encodes the temporal derivative of reward expectation (RPE). The study proposes a model where DA activity across distinct mesoaccumbal pathways can be conceptualized as state and rate-of-change encoding, analogous to what is observed in sensory systems. The results indicate that DA neurons exhibit parallel encoding of a state and a temporal derivative of value during motivated behavior. The study also highlights the importance of considering anatomical organization when interpreting DA signaling. The findings suggest that the brain uses a proportional-differential encoding mechanism for higher-order variables, such as state value, similar to sensory processing in the peripheral and central nervous systems. The study provides evidence that DA cell body activity and release are topographically organized along a dorso-lateral to ventro-medial gradient both at the cell body and terminal level. The results suggest that DA neurons in the mVTA are more likely to encode state than lVTA DA neurons, while lVTA DA neurons are more likely to encode the rate of change of state. The study also shows that DA release in the NAcMed lacks a negative RPE, making it unlikely that NAcMed DA release tracks a derivative-like signal. The findings suggest that DA signaling in the NAcMed contains additional information, such as general salience, aversion, novelty, or movement. The study provides new insights into the nature of the state vector encoded in mVTA DA neurons and in NAcMed DA release and how it contributes to behavior in higher-dimensional scenarios. The results open new research avenues about the role of DA in reward learning and motivated behavior. The study also highlights the importance of considering the anatomical organization of DA neurons when interpreting DA signaling. The findings suggest that DA neurons in the mVTA are more likely to encode state than lVTA DA neurons, while lVTA DA neurons are more likely to encode the rate of change of state. The study also shows that DA release in the NThe study explores how dopamine (DA) neurons in the medial and lateral ventral tegmental area (VTA) encode an animal's behavioral state and its rate of change during a reward-seeking task. It reveals that two distinct DA neuron subtypes encode either the current state or its rate of change. Medial VTA (mVTA) DA neurons show sustained activity, while lateral VTA (lVTA) DA neurons exhibit transient activity. These activity patterns align with DA release patterns in anatomically defined mesoaccumbal pathways. The findings suggest that the brain uses two parallel lines for proportional-differential encoding of a state variable and its temporal dynamics, similar to sensory systems. The study also shows that DA cell body activity directly translates to DA release within anatomically defined mesoaccumbal subsystems. Separate DA pathways encode reward expectation and reward prediction error (RPE). The medial nucleus accumbens (NAcMed) primarily encodes reward expectation, while the lateral nucleus accumbens (NAcLat) encodes the temporal derivative of reward expectation (RPE). The study proposes a model where DA activity across distinct mesoaccumbal pathways can be conceptualized as state and rate-of-change encoding, analogous to what is observed in sensory systems. The results indicate that DA neurons exhibit parallel encoding of a state and a temporal derivative of value during motivated behavior. The study also highlights the importance of considering anatomical organization when interpreting DA signaling. The findings suggest that the brain uses a proportional-differential encoding mechanism for higher-order variables, such as state value, similar to sensory processing in the peripheral and central nervous systems. The study provides evidence that DA cell body activity and release are topographically organized along a dorso-lateral to ventro-medial gradient both at the cell body and terminal level. The results suggest that DA neurons in the mVTA are more likely to encode state than lVTA DA neurons, while lVTA DA neurons are more likely to encode the rate of change of state. The study also shows that DA release in the NAcMed lacks a negative RPE, making it unlikely that NAcMed DA release tracks a derivative-like signal. The findings suggest that DA signaling in the NAcMed contains additional information, such as general salience, aversion, novelty, or movement. The study provides new insights into the nature of the state vector encoded in mVTA DA neurons and in NAcMed DA release and how it contributes to behavior in higher-dimensional scenarios. The results open new research avenues about the role of DA in reward learning and motivated behavior. The study also highlights the importance of considering the anatomical organization of DA neurons when interpreting DA signaling. The findings suggest that DA neurons in the mVTA are more likely to encode state than lVTA DA neurons, while lVTA DA neurons are more likely to encode the rate of change of state. The study also shows that DA release in the N