Psychiatry 74(1) Spring 2011 5 Why Does Depression Hurt? Panksepp and Watt Why Does Depression Hurt? Ancestral Primary- Process Separation-Distress (PANIC/GRIEF) and Diminished Brain Reward (SEEKING) Processes in the Genesis of Depressive Affect Jaak Panksepp and Douglas Watt What can affective neuroscience add to ry-process nature). Although every aspect of the discussion of the genesis of depression? the affective life can be influenced by depres- Among other contributions, it may begin to sion, depression is intimately related to 1) answer the question of why depression feels sustained overactivity of the separation-dis- so bad. Since it is the only basic neuroscience tress PANIC system that can, if prolonged, approach that specifically aims to take the lead to a downward cascade of psychological affective infrastructure of the evolved mind despair (a theoretical view originally formu- as its central focus, it offers testable hypoth- lated by John Bowlby); and 2) the despair eses concerning the affective imbalances that phase that follows the acute PANIC response contribute to clinical depression (Solms & which is characterized by abnormally low Panksepp, 2011). A critical question about activity of the SEEKING system. In terms genesis of depression is: Which negative of animal modeling, depression reflects the affect-generating networks of mammalian behavioral agitation of separation distress brains are most important for understanding followed by emotional shutdown. The initial depressive “pain” and what new therapeu- behaviorially agitated panic state may include tics might such knowledge engender? SEEKING arousal, followed by dramatically Affective neuroscience has outlined diminished SEEKING during the depressive seven primary process (i.e., genetically pro- “despair phase.” From this perspective, de- vided) emotional systems. All are subcorti- pression may have evolutionary advantages, cally situated (Panksepp, 1998), where ani- such as conservation of resources follow- mal models allow causal (vs. correlational) ing unalleviated separation distress. A more analysis, not afforded by human research, detailed exposition of this view, along with including modern brain imaging. These pri- seven expert commentaries, is available in mary functions consist of SEEKING, RAGE, Watt and Panksepp (2009). Here, we briefly FEAR, sexual LUST, maternal CARE, sepa- summarize the implications of this perspec- ration-distress PANIC/GRIEF (henceforth, tive for research and therapeutics. simply PANIC) and joyful PLAY (neural sys- Many stressors used to evoke depres- tems are capitalized to highlight their prima- sive phenotypes—from physical to psycho- Jaak Panksepp is affiliated with the Department of VCAPP, College of Veterinary Medicine, Washington State University, Pullman. Douglas Watt is with Cambridge City Hospital, Harvard Medical School, Cambridge, Mas- sachusetts. This work was supported by the Hope for Depression Research Foundation to JP. Address correspondence to Jaak Panksepp, Department of VCAPP, College of Veterinary Medicine, Washington State University, Pullman, WA 99164-6351. E-mail: [email protected]. © 2011 Guilford Publications, Inc. 6 Why Does Depression Hurt? logical (e.g., social-defeat in adult aggressive Depression research during the last encounters)—remain to be clearly linked four decades of the 20th century focused most to specific emotional network activities, al- heavily on the consequences of stress (DeK- though neuroanatomical correlates have loet, Joels, & Holsboer, 2005; McEwen, been identified (e.g., Kanarik et al., 2011). 2007) and brain norepinephrine and sero- Most animal models of depression not only tonin dynamics (from Schildkraut, 1965, to employ very general stressors but also gen- Harro & Oreland, 2001, so to speak). This eral behavioral outcome measures—various excellent work has largely neglected why de- timid behaviors and diminished pleasure re- pression feels so bad. It is becoming harder sponses (e.g., hopeless swimming patterns to believe that general brain serotonergic and diminished sexual eagerness). Rarely and/or noradrenergic changes, which glob- are specific emotional circuits prescribed as ally regulate aspects of forebrain arousal targets of analysis. Because of general stres- and dynamics, will specifically explain the sors and nonspecific outcome measures, such morbid moods of depression (Delgado et al., preclinical studies rarely provide insights 1990). These amines regulate quite general into psychodynamic or interpersonal consid- brain arousal functions that influence all erations of primary concern for clinical prac- emotions. It is no surprise that SSRIs mild- titioners, especially the feelings of social loss ly ameliorate many psychiatric problems, and defeat that promote depressive affects. while having modest overall efficacy, as Modern neuroscience, especially as applied highlighted by disappointing STAR*D find- to animal models, has little room for discus- ings (Rush, 2007; Rush, Trivedi, & Fava, sions of mental phenomena—especially the 2003). Likewise, more recent work on vari- affective experiences—that characterize psy- ous neurotrophic factor depletions (Koziek, chopathologies. Middlemas, & Bylund, 2008), stress-induced In line with Watt & Panksepp (2009) hippocampal shrinkage and CNS inflamma- and Solms & Panksepp (2011), we believe tion (Miller, Maletic, & Raison, 2009), and emotional-systems analyses will promote underlying genetics (Levinson, 2006), albeit better interdisciplinary dialog, yielding bet- of potential causal significance, provide little ter therapeutic interventions, where general understanding of the affective feelings that biogenic amine and cognitive-behavioral reg- characterize depression. Affective neurosci- ulatory strategies are commonly emphasized ence approaches can illuminate the subjec- more than affect-oriented therapies (but see tive manifestations of depressive affect (e.g., Fosha, Siegel, & Solomon, 2009, and Sh- Kroes et al., 2007; Panksepp, 2006; Pank- edler, 2010, for counter-examples). On the sepp et al., 2002.). psychopharmacological front, affective neu- roscience views promote the development AN AFFECTIVE NEUROSCIENTIFIC of new medicinals that target neuropeptide PERSPECTIVE ON WHY systems, such as endogenous opioids (both DEPRESSION FEELS SO BAD mu and kappa varieties) and corticotrophin releasing factor (CRF) dynamics, as well as amino acid transmitters, such as glutamate In extending the original Bowlby ac- and glycine, that substantively control many count of depression to depression neurobiol- affective states, including emotional learn- ogy (Watt & Panksepp, 2009), we here focus ing. On the psychotherapeutic front, affec- on how brain affective networks, altered by tive neuroscience approaches promote better sustained distress, may explain the psycho- and more specific utilization of positive emo- logical pain and dysphoria of depression. tions, such as facilitated SEEKING, CARE, Overactivity of brain separation-distress and PLAY dynamics, which currently remain PANIC/GRIEF and underactivity of SEEK- underdeveloped. ING networks may explain how the biggest Panksepp and Watt 7 FIGURE 1. Human and animal schematics of sadness and separation-distress systems. Animal data comes from localized brain stimulation mapping of separation distress circuits in guinea pigs (Herman & Panksepp, 1981) and human data from Damasio, et al., 2000 (from Panksepp 2003). epidemiological stressor, namely social loss, the last two PANIC facilitators have yielded promotes depression (Bowlby, 1980; Heim promising antidepressant effects (Holsboer, & Nemeroff, 1999). Depression may feel bad 2000; Zarate et al., 2006). Likewise, it is to because brain separation-distress systems be anticipated that opioid and oxytocin fa- create psychological pain. As Bowlby recog- cilitators may alleviate depression. Indeed, nized, separation distress—the “protest” or opioids were widely used as antidepressants “panic” responses that promptly follow so- before the modern era (Tenore, 2008), and cial loss, especially in young animals—feels ultra-lowdoses of buprenorphine are fine an- bad in a unique way. Affective understanding tidepressants for individuals not helped by of such brain processes, garnered by map- traditional medications (Bodkin et al., 1995). ping neuroanatomies and neurochemistries Perhaps oxytocinergic and prolactinergic of separation distress, helps clarify the na- drugs can be harnessed for similar ends. ture of social attachments and loss (Pank- sepp, 1998). These neuroanatomies are sum- BUT SEparaTION DISTRESS marized in Figure 1. IS ONLY THE GATEwaY TO The PANIC/GRIEF circuitry starts in DEPRESSION midbrain central gray regions, commonly called the periaqueductal gray (PAG), and ascends through medial diencephalic struc- The PANIC/GRIEF system prob- tures, especially the dorsomedial thalamus, ably evolved from general pain mechanisms and terminates in various basal forebrain (Panksepp, 1998, 2003) presumably millions nuclei and subcallosal anterior cingulate of years ago (birds possess a homologous forebrain regions, which have been targeted system). This psychic-pain system promotes for deep brain stimulation (DBS) therapy for social cohesion, forges bonds between in- treatment-resistant depressions (Mayberg et fants and caretakers, and fortifies friendships al., 2005). Key neurochemistries that pro- and sexual relationships—in short, promotes mote separation calls (protest) are declining social solidarity among group living