this discusses a lot of them: https://web.hansonwadegroup.com/rs/355-DOS-429/images/9th%20Neuropsychiatric%20Drug%20Development%20Summit%202026%20brochure.pdf
Salamone’s central idea is that mesolimbic dopamine, especially nucleus accumbens dopamine, is less a pleasure chemical than a controller of behavioral activation, vigor, persistence, and willingness to pay effort costs. In his animal paradigms, rats choose between something like:
press a lever repeatedly for the preferred food
versus
eat mediocre chow sitting right there.Reduce dopamine function and the rat often still likes/wants food, but switches toward the low-effort option. That gives you a model of anergia/apathy without necessarily producing anhedonia. This framework has been applied to depression, schizophrenia, Parkinson’s disease, inflammatory states, etc. (PubMed Central (PMC))
The “atypical dopamine transport inhibitors” bit is particularly interesting
He’s likely discussing his group’s work on modafinil-like DAT inhibitors that raise dopamine without behaving exactly like cocaine or conventional stimulants.
Their recent paper tested:
JJC8-088, JJC8-089, RDS3-094 and JJC8-091.
Three of them reversed a tetrabenazine-induced low-effort state in rats, while JJC8-091 didn’t, apparently related to the unusual conformation of DAT that it stabilizes. JJC8-088 and JJC8-089 also increased willingness to work under progressive-ratio conditions. (PubMed)
The therapeutic dream is roughly:
$$
\text{DAT inhibition}
\rightarrow \uparrow DA_{\text{accumbens}}
\rightarrow \downarrow \text{subjective effort cost}
\rightarrow \uparrow \text{goal-directed action}
$$but without
$$
\text{euphoria / compulsive reinforcement / psychosis / high abuse liability}.
$$Earlier compounds in this line include modafinil, (S)-CE-123, CT-005404 and (S)-MK-26; his group has also shown pro-effort effects from bupropion, methylphenidate and lisdexamfetamine in these paradigms. (PubMed)
And the “non-dopaminergic” part probably means A2A in particular
This is a long-running Salamone theme:
$$
\boxed{\text{Adenosine A}_{2A} \leftrightarrow D_2\text{ dopamine signaling}}
$$A2A and D2 receptors interact strongly in striatal circuitry. A2A antagonism tends to push behavior toward greater effort expenditure, and in his animal models it can reverse motivational impairment caused by dopamine antagonism, tetrabenazine, or inflammatory cytokines. (PubMed Central (PMC))
The clinically recognizable A2A antagonist is istradefylline, approved for Parkinson’s disease. Caffeine is the crude everyday version because it antagonizes both A1 and A2A receptors, although it isn’t remotely as selective. (Neuroscience Blueprint)
And they may go further than adenosine. A 2025 study explicitly compared the atypical DAT inhibitor CE-123 with NMDA-receptor blockade in a tetrabenazine model of apathy, so glutamatergic mechanisms are also floating around this program of research. (PubMed)
The deeper conceptual point
I suspect this will be the most interesting part of the talk.
They’re probably arguing against treating “motivation” as one scalar variable.
Someone can have:
- normal consummatory pleasure: “This would be enjoyable.”
- normal reward valuation: “I want that outcome.”
- intact knowledge of what to do
- but abnormally high effort discounting: “It isn’t worth initiating the actions required to get there.”
Formally, imagine action value as something like
$$
V_{\text{action}}R-\lambda E
$$where (R) is expected reward and (E) is effort cost.
Apathy could arise because (\lambda) is too large rather than because (R) has collapsed.
That distinction matters enormously for treatment. An SSRI might improve negative affect yet do almost nothing to the pathological effort term. In fact, Salamone’s animal work is one reason people have become interested in dopaminergic/pro-effort treatments for residual fatigue and anergia, rather than assuming another serotonergic antidepressant will solve it. (PubMed)
So the title’s “integrated framework” probably means something close to:
$$
\boxed{
\text{Apathy}
\neq
\text{simply low dopamine}
}
$$Instead, identify which motivational computation is broken, then hit complementary mechanisms such as DAT/dopamine + A2A/adenosine + potentially glutamatergic or inflammatory pathways.
And this isn’t me guessing the conference: I found the actual session. It’s John Salamone’s September 17, 2026 talk at the 9th Neuropsychiatric Drug Development Summit in Boston, and the published agenda contains exactly the passage you pasted. (Neuropsychiatric Summit 2026)
Given your interest in ADHD / task initiation specifically, there’s a very interesting connection here: Salamone’s framework gives a much better vocabulary for distinguishing “I don’t value this” from “I cannot mobilize enough activation to cross the action threshold.” Those can look identical from the outside while being computationally quite different.