Few topics in contemporary trauma research are as fascinating — and as prone to overinterpretation — as the hypothesis that trauma is biologically transmitted to the next generation. For genogram work, the central question is: When family symptoms stubbornly repeat across generations, what is relationship history, what is narrative, what is biology? This lesson sorts out the levels, acknowledges the findings, and states very clearly where the evidence ends.

Three Levels of Transgenerational Transmission

Those who speak of "transgenerational trauma" almost always conflate three levels that should be analytically separated. The psychological level encompasses everything transmitted via attachment experience, parental mirroring, early relational patterns, and unconscious identifications. This is the oldest and best-documented pathway — from Selma Fraiberg’s "Ghosts in the Nursery" (1975), through attachment research, to contemporary mentalisation studies. The social level includes narratives, silences, family myths, economic consequences (lost wealth, interrupted education), living conditions, and communities. Here too, the evidence is robust: Holocaust survivors and their descendants, war children, children of politically persecuted parents all show measurable biographical and epidemiological effects across at least two generations.

The biological level is the most recent, methodologically demanding, and simultaneously the most popular. It comprises two sub-hypotheses: First, changes in maternal physiology during pregnancy and breastfeeding that shape the child’s development (intrauterine pathway). Second — and here it becomes speculative — the direct transmission of epigenetic marks via egg or sperm, such that even grandchildren who never had contact with their traumatised grandparents are supposed to bear biological traces. Those who do not clearly separate these levels risk presenting popular science simplifications ("trauma is inherited in the genes") as established fact.

Epigenetics Basics Without Mystification

Epigenetics refers to biochemical modifications on or around DNA that regulate gene activity without altering the base sequence itself. The two most important mechanisms are DNA methylation (attachment of a methyl group to cytosine bases, usually in the promoter region, which generally dampens gene expression) and histone modifications (chemical changes to the proteins around which DNA is wrapped, making DNA more or less accessible to transcription enzymes). These mechanisms are established molecular biology — they are responsible for the fact that different cell types arise from a fertilised egg, even though all contain the same DNA.

A crucial property for the trauma discussion is this: During germline formation and shortly after fertilisation, the epigenome in mammals is largely reprogrammed — methylation patterns are erased and reset. For an experience-dependent epigenetic mark to actually be passed on to grandchildren, it would have to survive this double reprogramming barrier. That this occurs in humans is not convincingly demonstrated — a point often omitted in popular accounts. When reading about "epigenetic inheritance of trauma", one should ask whether "inheritance" in the strict sense (via the germline, to F2 and beyond) or merely "intergenerational transmission" (to F1 via intrauterine effects) is meant. Popular language blurs the distinction — science cannot afford to do so.

The Yehuda Studies: Groundbreaking and Limited

Rachel Yehuda’s group at Mount Sinai (New York) has shaped the field. The 2016 study published in Biological Psychiatry examined methylation patterns at the FKBP5 gene — a gene that regulates stress response sensitivity via the glucocorticoid receptor — in 32 Holocaust survivors and their adult children, compared to a demographically matched control group. The finding: parents and children showed changes at a specific FKBP5 region (intron 7), but in opposite directions — a remarkable constellation.

The result made international headlines — and was simultaneously methodologically thin. The sample of 32 family pairs is small for epigenetic studies, where effects are typically weak and masked by many confounders. Lifestyle factors (diet, smoking, personal stress, socioeconomic status of the subsequent generation) were not fully controlled. Replication is still pending. Yehuda herself has repeatedly contextualised the findings with appropriate caution — the media reception often did not. Later work by the group on Syrian refugee families (2014/2015) and on cortisol profiles in pregnant women after September 11th follows the same pattern: methodologically careful enough to be taken seriously, too small to allow definitive conclusions.

The Critical Response: Horsthemke 2018

The most important critical counterpoint is Bernhard Horsthemke’s review article "A critical view on transgenerational epigenetic inheritance in humans" (Nature Communications 2018). Horsthemke’s argument comes down to a single, precise point: What is presented in the Holocaust literature and popular reporting as "transgenerational" inheritance is, upon closer inspection, in almost all cases intergenerational — that is, transmitted to F1 via maternal physiology during pregnancy or via early postnatal interaction. This would be highly relevant for clinical practice, but not surprising. For true transgenerational inheritance (to F2 or F3, to grandchildren or great-grandchildren, without these ever having direct contact with the traumatic event or with the pregnant maternal ancestors), there is, according to Horsthemke, no convincing evidence in humans.

This distinction is not academic nitpicking. Advising a third-generation (G3) client that she carries epigenetic trauma marks from her grandparents is to claim something not supported by the data. What she likely carries are parenting, attachment, and narrative patterns from her mother (G2), who herself was the child of a traumatised mother — a fully psychosocial causal chain that requires no additional biological hypothesis. The DZPG research programme (German Centre for Mental Health, since 2023) has made this differentiation its agenda and invests in rigorous longitudinal designs that distinguish intergenerational from transgenerational effects.

Animal Models and Their Limits

Animal studies are frequently cited in trauma epigenetics. Dias & Ressler (2014) conditioned male mice to a smell (acetophenone) using electric shocks and found that their offspring and grand-offspring showed increased sensitivity to exactly this smell, as well as changes in olfactory brain regions — in animals that had never themselves been conditioned. Eric Nestler’s group (Mount Sinai) has shown similar findings in mouse models of chronic social stress. These studies are methodologically elegant and support the principle possibility of transgenerational epigenetic effects via the germline.

However, extrapolating to humans is not straightforward. Mice have shorter generation times, different germline reprogramming patterns, and above all, no language, no elaborate narrative communities, no schools, and no institutional history. In humans, psychosocial transmission pathways are so dense and effective that biological effects are methodologically almost impossible to disentangle. As experienced replication researcher Kevin Mitchell has put it: Even if epigenetic inheritance does exist in humans, its contribution to clinical symptomatology would likely be small compared to the impact of psychosocial and early physiological pathways.

Neurobiological Correlates: HPA Axis and FKBP5

What is well established and clinically relevant are altered neuroendocrine stress profiles in the offspring of severely traumatised parents. Central here is the hypothalamic-pituitary-adrenal axis (HPA axis) and its main messenger, cortisol. Yehuda and others have shown in several studies that adult children of Holocaust survivors with parental PTSD exhibit abnormal cortisol profiles — often in the direction of increased glucocorticoid sensitivity, not necessarily elevated cortisol levels per se. The FKBP5 gene encodes a protein that regulates glucocorticoid receptor sensitivity; FKBP5 polymorphisms interact with childhood trauma in predicting later PTSD symptoms (Binder et al., classic work).

Two further neurobiological findings are part of the picture: Increased reactivity of the amygdala (central brain structure for threat processing) in the offspring of traumatised parents can be demonstrated in imaging studies — effect sizes are moderate, methodological heterogeneity is high. Altered prefrontal top-down control of the amygdala is also discussed. However, these findings are likewise open to purely psychosocial explanations: Growing up in an environment with chronically hyper-alert parents calibrates the threat system accordingly — without requiring a separate biological inheritance hypothesis. Clinically, this is often the more sober and no less respectful interpretation.

Consequences for Genogram Work

From this state of research, three very practical consequences follow for genogram work. First: It is legitimate and meaningful to explicitly mark in the genogram when G1 or G2 have experienced severe collective traumas — Holocaust, war, displacement, political persecution, colonial violence. Omitting this deprives the client of an important context for their own symptoms. Second: It is methodologically and ethically problematic to causally interpret a client’s symptom (e.g., diffuse anxiety, hyperarousal, dissociative tendencies) as an "epigenetic trauma mark of the grandmother". The evidence does not support this specificity, and the message can be iatrogenic — it suggests a biological determinism that is neither scientifically justified nor therapeutically helpful.

Third: The productive clinical message is a different and more respectful one. It is roughly: "Your experience has a context. Your family’s history is part of that context. Which memories, narratives, silences and response patterns you have taken on — that is material for our work. Whether there are also biological traces running alongside, we cannot say for sure — but for your healing, it is more important to understand how you now deal with this legacy than to trace the exact source of your stress reactivity to a methyl group or a family story." This stance is trauma-sensitive in the best sense: It takes family history seriously without falling into deterministic reductionism.

Practice Vignette

A 41-year-old client, granddaughter of a maternal Holocaust survivor, comes to the practice with diffuse jumpiness, sleep disturbances, and a chronic feeling of "never being quite safe". She has no trauma experience of her own in the narrow sense. In previous therapies, she encountered popular science texts on epigenetically inherited trauma and developed a self-concept from them: "I am biologically damaged by what happened to my grandmother."

In the genogram, the grandmother’s persecution history is carefully recorded and acknowledged — as a real, formative context for the family across three generations. In conversation, however, a distinction is made: The mother (G2) was raised by a chronically hyper-alert, ambivalently attached mother, and passed on this hyperarousal configuration in her own parenting — the client learned as a child to be quiet, "not to provoke", to respond instantly to her mother’s mood changes. This psychosocial transmission chain fully explains her symptoms. The message to the client is not "You are genetically damaged", but "You learned in a system that itself was still under stress — that is understandable, and it is changeable." As therapy progresses, the self-labelling as "biologically damaged" disappears and is replaced by more agentic interpretations.

State of Research & Discussion

The serious state of research at the end of 2025 can be summarised in five points: First, intergenerational effects of severe trauma on F1 are robustly documented — at the level of attachment, symptomatology, epidemiological outcomes, and cortisol profiles. Second, a biological mediation of these intergenerational effects via intrauterine and early postnatal mechanisms is plausible and increasingly empirically supported. Third, true transgenerational (F2+) epigenetic inheritance in humans remains hypothetical — animal data support the possibility, but human data are insufficient for robust conclusions. Fourth, FKBP5 and the HPA axis are serious research objects with clinical relevance; individual methylation findings from small samples should not be overinterpreted.

Fifth — and this is the most important message for practice: Even if all optimistic hypotheses about biological inheritance were confirmed, the clinical consequence would be less spectacular than expected. Trauma-sensitive relational work, stabilisation, working through family narratives, biographical integration — these tools work just as well, regardless of whether an additional biological pathway is present or not. The clinical response to transgenerational trauma is not epigenetic knowledge, but epigenetic humility: We know less than popular texts suggest, and we need to know less than they claim in order to be effective helpers.

A second aspect of practical consequence concerns the language in which clients think about themselves. Popular science accounts of transgenerational epigenetics — magazine articles, non-fiction books, TED talks — have, over the past ten years, become a significant reservoir for self-interpretation. Clients come with statements like "I inherited this", "It’s in my genes", "This is my grandparents’ burden". Such statements are not simply "wrong" — they carry a biographical truth that deserves to be taken seriously. But they also carry a biological claim that the evidence does not support, and they can tip into a deterministic self-perception that undermines the changeability of one’s own situation. A careful therapeutic response honours the biographical content of these statements and gently differentiates the biological claim — not in a lecturing mode, but in a joint search for more precise words for what the client is really trying to describe.

One final methodological note: The replication crisis in epigenetics is part of the broader replication crisis in the life sciences. Methylation studies are particularly vulnerable — sample sizes are often too small, statistical corrections for thousands of loci examined in parallel are demanding, confounders from age, cell type, sex, and lifestyle are substantial. An experienced reviewer of "Epigenome-Wide Association Studies" (EWAS) like Caroline Relton has repeatedly pointed out that the field has only developed the necessary methodological standards for robust findings since about 2020. Anyone reading current studies on trauma epigenetics should therefore ask: How large was the sample, were lifestyle confounders controlled, was the study independently replicated? If not, the finding remains interesting, but provisional.

Cross-References

• M3 "Trauma-Sensitive Practice" — the clinical foundation on which biological knowledge builds, not replaces.
• M3 "Bowen Differentiation" — Bowen’s concept of transgenerational transmission was pioneering and does not rely on biological hypotheses.
• M4 Lesson 6 — chronic illness and genetic inheritance anxiety, methodologically related.
• M4 Lesson 19 — refugees, acute multi-generational trauma constellations.
• M4 Lesson 17 — reliability and self-critique, applies especially strictly to biological hypotheses.

Further Links

• https://www.nature.com/articles/s41467-018-05445-5 (Horsthemke 2018 — the methodologically rigorous critical response, essential reading)
• https://www.dzpg.org/en/latest-news/press-releases/article/80-years-after-the-end-of-the-war-dzpg-research-explores-the-intergenerational-impact-of-trauma (DZPG research programme)
• https://www.apa.org/monitor/2019/02/legacy-trauma (APA overview, popular science but carefully differentiated)
• https://www.nature.com/articles/s41598-025-12085-5 (current study on G3 Holocaust survivors, read critically)
• https://www.research.va.gov/currents/1016-3.cfm (Yehuda in the US veteran context)