Charitable Estate Planning as Visualized Autobiography

Behavioral

Converting the projected likelihood of making a charitable bequest into numbers as 1 for “None,” 2 for “Unlikely,” 3 for “Somewhat Likely,” and 4 for “Highly Likely” resulted in an average of 2.09 for bequest questions, 2.22 for giving questions, and 2.42 for volunteering questions. Only 27.9% of all bequest responses were in the “Somewhat Likely” or “Highly Likely” categories, as contrasted with 39.5% of all giving responses, and 45.7% of all volunteering responses. The response time to answer the bequest questions was also greater, overall, than for the other types of questions (see Table 1).

OPEN IN VIEWER

Table 1.

Self-Reported Likelihood to Give, Volunteer, or Bequest.

Category	(1) “None”	(2) “Unlikely”	(3) “Somewhat likely”	(4) “Highly likely”	Missing	Average response
Bequest	30.7%	38.9%	16.6%	11.3%	2.5%	2.09
Give	30.5%	28.3%	26.8%	12.7%	1.8%	2.22
Volunteer	24.4%	29.1%	25.8%	19.9%	0.8%	2.42
Bequest response time	3.49	4.08	3.42	3.57		3.74
Give response time	3.03	4.08	3.92	3.49		3.64
Volunteer response time	2.69	3.73	3.45	2.52		3.18

fMRI

Table 2 reports areas of differential activation with charitable bequest blocks as contrasted with charitable giving blocks, volunteering blocks, and charitable giving and volunteering blocks combined. Only areas significant in either the peak-level or cluster-level at p < .05 after correction for family-wise error (FWE) are reported. The first three contrasts reflect areas of greater activation during the charitable bequest condition. These results consistently show lingual gyrus and precuneus activation in contrasts with giving, volunteering, and giving and volunteering combined. The only additional result to pass the p < .05 FWE threshold was a primary motor cortex (precentral gyrus) activation in the volunteering contrast. This primary motor cortex difference is likely due to differences in button pushes ² between the two conditions (Cramer, Finklestein, Schaechter, Bush, & Rosen, 1999; Yousry et al., 1997) and is not of interest for the present analysis.

OPEN IN VIEWER

Table 2.

Relative Activations Comparing Charitable Bequest With Giving and/or Volunteering (Reporting Only p < .05 FWE Corrected).

			Peak-level		Cluster-level
Contrast	Title	MNI Coordinates	p (FWE-corr)	Z-score	p (FWE-corr)	k e
1. Bequest>give	Lingual Gyrus	−2, –78, –2	0.004	5.44	0.000	1399
	Precuneus	26, –66, 42	0.102	4.64	0.009	313
2. Bequest>volunteer	Lingual Gyrus	2, –80, –4	0.007	5.32	0.000	2254
	Precuneus	30, –66, 40	0.180	4.47	0.004	356
	Precentral Gyrus	−34, –3, 36	0.397	4.19	0.001	433
3. Bequest>(give or volunteer)	Lingual Gyrus	0, –78, –4	0.001	5.82	0.000	2016
	Precuneus	26, –66, 42	0.007	5.33	0.001	475
4. Give>bequest	Cuneus	8, –88, 20	0.157	4.51	0.003	388
5. Volunteer>bequest	Cuneus	8, –88, 20	0.060	4.79	0.000	1008
	Insula	−38, –20, 8	0.323	4.27	0.001	462
6. (Give or Volunteer)>bequest	Cuneus	6, –90, 22	0.011	5.22	0.000	1014

Note: Height threshold T = 3.73 (p < .001 uncorrected); Extent threshold (k = 0 voxels); Voxel size 2 × 2 × 2 mm; Volume 184,660 voxels; Expected voxels per cluster, k = 22.45 (C1); 21.982 (C2); 22.706 (C3); 22.45 (C4); 21.982 (C5); 22.706 (C6); FWHM (in mm) = 14.7 × 14.5 × 11.9 (C1); 14.6 × 14.3 × 11.9 (C2); 14.8 × 14.5 × 12 (C3); 14.7 × 14.5 × 11.9 (C4); 14.6 × 14.3 × 11.9 (C5); 14.8 × 14.5 × 12 (C6).

In the converse analysis, the cuneus displayed consistently greater activation for either or both charitable giving and volunteering questions compared with bequest questions. The sole remaining difference was a greater activation in the insula with volunteering questions that was significant at the cluster-level, but not peak-level. Figure 1 displays the location of the significant differential activations comparing bequest decisions with the combination of both giving and volunteering decisions.

OPEN IN VIEWER

Figure 1.

Brain image of differential activations in bequest as compared with giving or volunteering.

Where the contrasts in Table 2 sought to discover neurological differences between charitable bequest decisions and other types of charitable decisions, Table 3 reports activations associated with a higher or lower predicted likelihood of making a charitable bequest. Once again, we set aside from consideration the lateralized results from the precentral and postcentral gyrus as likely reflecting the association between physical button pushes and agreement level. This results in only two areas of interest with significant activation. Lingual gyrus activation was associated with higher projected likelihood of making a charitable bequest. Conversely, insula activation was associated with lower projected likelihood of making a charitable bequest.

OPEN IN VIEWER

Table 3.

Activating With Increasing and Decreasing Charitable Bequest Agreement (Linear Parametric Modulation Reporting Only p < .05 FWE Corrected).

			Peak-level		Cluster-level
Contrast	Title	MNI coordinates	p (FWE-corr)	Z-score	p (FWE-corr)	cluster size
1. Increasing with agreement	Lingual Gyrus	10, –68, –4	0.004	5.46	0.000	671
	Postcentral Gyrus	−40, –22, 52	0.007	5.37	0.000	1200
2. Increasing with disagreement	Precentral Gyrus	38, –20, 62	0.000	6.20	0.000	1387
	Insula	42, –20, 18	0.171	4.61	0.013	196

Note: Height threshold T = 3.73 (p < .001 uncorrected); Extent threshold (k = 0 voxels); Voxel size 2 × 2 × 2 mm; Volume 184,660 voxels; Expected voxels per cluster, k = 13.976; FWHM (in mm) = 12.5 × 12.3 × 10.2.

Taking an Outside Perspective on One’s Self

The precuneus appears to be differentially involved in taking the viewpoint of observing one’s self from an outside perspective (Vogeley & Fink, 2003) and has been referred to as the “mind’s eye” (Fletcher et al., 1995). Kjaer, Nowak, and Lou (2002) found greater precuneus activation when participants described their own physical characteristics and personality traits as compared to describing the physical characteristics and personality traits of another person. Lou et al. (2004) compared changes when recalling judgments of psychological traits of one’s self, one’s best friend, and a neutral reference person. Precuneus activation was greatest when referencing one’s self, moderate when referencing one’s best friend, and lowest when referencing a neutral reference person. Correspondingly, transcranial magnetic stimulation disrupting normal neural circuitry in this area slowed the ability to recall judgments about one’s self significantly more than the ability to recall judgments about others. Summarizing a wide range of such studies, D’Argembeau et al. (2007, p. 935) wrote, “The neural correlates of self-referential processing most consistently observed across these studies are located in the medial prefrontal cortex and in medial posterior regions (in the posterior cingulate cortex or the precuneus)”

D’Argembeau et al. (2007) compared first person perspective with third person perspective judgment by contrasting participants’ own opinions about self and another person with participants’ projections about the other person’s opinions about self and the other person. The most significant area of activation associated with third person perspective as contrasted with first person perspective was the lingual gyrus. (The precuneus was the fourth most significantly associated region). The lingual gyrus has commonly been associated with taking a third person perspective. Even in simple motor tasks, creating a third person as contrasted with a first person perspective generates significantly greater activation in the lingual gyrus (Hesse, Sparing, & Fink, 2009; Jackson, Meltzoff, & Decety, 2006)

The suggestion that taking a third person perspective is inherent in a bequest task makes sense considering the timing of the transfer, that is, after death. It is easy to imagine personally giving a current charitable gift or volunteering time to a cause. However, imagining a bequest transfer is different as, at the time of transfer, the actor is deceased. If requests for bequest giving invoke images of one’s status of being deceased (i.e., death saliency), it is easy to see how such contemplation would involve observing one’s self (perhaps one’s deceased self) from an outside perspective.

Death and Grief

There have been relatively few fMRI studies on the topic of death-related emotion. ³ Gündel, O’Connor, Littrell, Fort, and Lane (2003) worked with participants who had lost a first-degree relative in the previous year. Participants were interviewed about the death, the cause of death, and the memorial service for the deceased. From this interview, 15 grief-related words with an autobiographical connection to the death of the loved one (such as “collapse,” “funeral,” or “loss”) were contrasted with neutral words, both with and without a background picture of the deceased. The only region showing significant activation (at p < .05, corrected) in response to grief-related words (as contrasted with neutral words) was the precuneus. This may be particularly relevant for the present study, as the topic of the bequest questions were presented only in text. The precuneus did not differentially activate for the picture of the deceased; however, one of only two significant (at p < .05, corrected) activation areas was the lingual gyrus. ⁴

O’Connor et al. (2008) followed a similar protocol but reported results only for specific pain and reward regions related to a specialized diagnosis of “complicated grief” and thus excluded our regions of interest. No other fMRI studies have been conducted on grief from the loss of a human loved one (O’Connor, Gündel, McRae, & Lane, 2007, reanalyzed the same fMRI data as in Gündel et al., 2003, but for different purposes). However, Freed, Yanagihara, Hirsch, and Mann (2009) examined grief caused by the loss of a pet. Twenty participants who had lost a pet dog or cat within the previous 3 months were presented with words previously identified to remind them of their deceased pet. Of the 12 areas showing significant activity in response to the deceased reminder words, 4 were located in the precuneus.

Autobiographical Memories

Reminders of a recently deceased loved one activated the precuneus for words (Freed et al., 2009; Gündel et al., 2003) and the lingual gyrus for images (Gündel et al., 2003). These two areas were also prominent in a study of autobiographical family memories in Gilboa, Winocur, Grady, Hevenor, and Moscovitch (2004). Participants (average age of 51) were presented with photographs from across their life, corresponding with ages 5 to 11, 11 to 20, 20 to 30, 30 to 40, and after 40. The photographs were selected by family confederates without the participant’s involvement. Each photograph depicted some event (not simply portraits) and included the participant. Precuneus and lingual gyrus activation occurred when the participants were able to vividly recall and relive the event in the picture, but not where the scenes were only vaguely familiar. The authors explained, “retrieving detailed vivid autobiographical experiences, as opposed to personal semantic information, is a crucial mediating feature that determines the involvement of hippocampus and two posterior neocortical regions, precuneus and lingual gyrus, in remote autobiographical memory.” (Gilboa et al., 2004, p. 1221)

In a similar study with participants 60 to 70 years of age, Viard et al. (2007) had participant family members identify vivid events that had occurred during the ages of 0 to 17, 18 to 30, 31 to previous 5 years, previous 5 years to previous year, and previous year. During the experiment, participants were presented with text reminders of each event and asked “to mentally relive personal episodes . . . by ‘traveling back in time’ and remembering as much detail as possible” (Viard et al., 2007, p. 2455). Regardless of the recency of the event, such reliving differentially activated the precuneus and lingual gyrus. Of the six regions showing significant activation, two were located in the precuneus and two in the lingual gyrus. In Denkova (2006), three of the four most statistically significant regions associated with recalling autobiographical personal events were in the lingual gyrus and precuneus areas.

In a study of autobiographical memory with Alzheimer’s patients and nonimpaired controls, Meulenbroek, Rijpkema, Kessels, Rikkert, and Fernández (2010) had participants answer true-false statements about their most salient lifetime events gathered from a previous interview. For both Alzheimer and control groups, the most statistically significant region of activation for the autobiographical memory tasks when contrasted with generic true-false semantic questions was the precuneus. However, Alzheimer’s patients showed enhanced activity in four regions including the right precuneus and left lingual gyrus as compared with control participants.

In a study contrasting autobiographical memory with third person perspective, Rabin, Gilboa, Stuss, Mar, and Rosenbaum (2010), suggested that among other areas, the “bilateral precuneus may respond more strongly to familiar events involving the self and possibly when the self is projected across time.” D’Argembau et al. (2007) summarized,

Finally, the precuneus and the left temporal pole have also been observed to be activated in earlier perspective-taking studies (Ruby & Decety, 2001, 2003, 2004) and this has been related to imagery, autobiographical memory retrieval and semantic processing (Cabeza & Nyberg, 2000). (p. 941)

D’Argembau et al. (2007) further suggested, “activation of the visual cortex (in the lingual gyrus) might also be related to autobiographical memory retrieval and in particular to visual imagery components, which play a key role in autobiographical memory (Greenberg & Rubin, 2003)” (p. 941). Taken together, these suggest that the areas of activation in the present study are also associated with autobiographical visualizations.

Deactivations

The cuneus was significantly deactivated during bequest-giving questions as contrasted with charitable giving and volunteering questions. This result is also consistent with the concept that bequest decision making is unique in mandating a third person perspective. (Again, it is easy to imagine writing a check or volunteering at a school from a first person perspective, but postmortem occurrences are inconsistent with first person perspective because the self is no longer an actor). Jackson et al. (2006) found both that the lingual gyrus was associated with third person perspective and simultaneously that the cuneus was associated with first person perspective. Two of the three areas of significantly greater activation while observing the first person perspective were located in the cuneus as were three of the seven areas of greater activation in imitating a first person perspective (Jackson et al., 2006). Similarly, Wurm, von Cramon, and Schubotz (2011) found greater activation in the lingual gyrus for third person perspective and simultaneously greater activation in the cuneus for first person perspective. Others have also associated cuneus activity with first person perspective-taking as contrasted with third person perspective-taking (David et al., 2006; Lorey et al., 2009). In the present case, cuneus activation was greater with giving and volunteering questions than bequest questions. This may relate to the ease with which giving cash or volunteering can be imagined in the first person, whereas bequest transfers are necessarily executed by someone other than the deceased.

In the charitable bequest questions, insula activation had a negative relationship with the self-reported likelihood of making a bequest. However, insula activation was also associated with volunteering decisions as contrasted with bequest decisions. A wide range of emotional functions engage the insula, including actual and social pain, but also happiness, sadness, disgust, fear, and anger (Eisenberger, Lieberman, & Williams, 2003; Rainville, 2002). Thus it is plausible that one type of emotion is associated with rejecting the charitable bequest (perhaps social discomfort from being viewed as not generous or disgust at being asked), but that a different type of emotion is associated with insula activation for volunteering questions.

The Visual Autobiography

The activation of both the lingual gyrus and the precuneus indicate that the processing taking place is primarily related to internally generated imagery. The lingual gyrus in the occipital lobe is part of the visual system and damage to the lingual gyrus can result in losing the ability to dream (Bischof & Bassetti, 2004). The precuneus has been called “the mind’s eye” and is required for visual imagery of memories (Fletcher et al., 1995).

Creating images for the mind’s eye of a desired outcome is a good approach when working with clients or donors in a variety of situations. However, in this case, knowing that the distinctive cognitive processing being used is clearly imagery-based heightens the importance of helping the client to “paint a picture” of the desired outcome. No doubt, complex planning must involve the requisite tax calculations and legal structures, but the core decision of selecting recipients appears to be one of visual imagery.

Beyond being visual, the results suggest that the bequest decision may be analogous to creating the final chapter in one’s personal autobiography (see also Schervish, 2006). The precuneus and lingual gyrus activation (as well as the cuneus deactivation) are consistent with the idea that participants are taking an outside or third person perspective looking back on themselves during bequest decision making. The precuneus and lingual gyrus are also engaged during the recollection of vivid autobiographical memories from across the entire life span. As such, it might be useful to think of the goals in estate planning or planned giving in an autobiographical/narrative form.

Routley (2011) also identified the importance of autobiographical connection when interviewing donors with planned bequests writing, “Indeed, when discussing which charities they had chosen to remember, there was a clear link with the life narratives of many respondents” (p. 220). Some examples from the interviews in Routley (2011) illustrate the sense in which charitable bequest decisions are part of one’s autobiography:

[In my will] there’s the Youth Hostel Association, first of all . . . it’s where my wife and I met . . . Then there’s the Ramblers’ Association. We’ve walked a lot with the local group . . . Then Help the Aged, I’ve got to help the aged, I am one . . . The there’s RNID because I’m hard of hearing . . . Then finally, the Cancer Research. My father died of cancer and so I have supported them ever since he died. (Male, 89, married)

The reason I selected Help the Aged . . . it was after my mother died . . . And I just thought—she’d been in a care home for probably three or four years. And I just wanted to help the elderly . . . I’d also support things like Cancer Research . . . because people I’ve known have died . . . An animal charity as well . . . I had a couple of cats. (Female, 63, widowed, pp. 220-221)

Estate planning clients and planned giving donors may, in a sense, be creating a story—a story that is part of one’s life story. To the extent that current plans (or lack of plans) are inconsistent with that life story, advisors and fund-raisers may be able to raise sufficient cognitive dissonance to spur action. This inconsistency may come from planning (or lack of planning) that causes a divergence from, for example, a lifetime of charitable actions, or perhaps a lifetime of avoiding unnecessary taxes and expenses. Identifying potential negative results that contradict one’s desired autobiographical identity, or positive possibilities that enhance one’s desired autobiographical identity, may be key to motivating action. Furthermore, fund-raisers may consider emphasizing the autobiographical connections between the donor and the charity, rather than focusing on the charity’s need for funds.

The Management of Death Salience

Bequest decision-making processes differentially activated areas similar to those involved in using death-oriented words to evoke memories of a recently deceased loved one (Freed et al., 2009; Gündel et al., 2003). This association is consistent with the rather obvious idea that bequest decisions involve reminders of mortality. Although such a suggestion is unlikely to produce much controversy, it can lead to practically useful implications due to the large body of research on the effects of mortality salience.

In psychology, “terror-management theory” suggests two levels of reactions or “defenses” to mortality salience (Pyszczynski, Greenberg, & Solomon, 1999). The initial defenses, labeled as proximal defenses, are focused on removing death reminders from awareness (Hirschberger, 2010). This can involve a variety of avoidance strategies such as denying one’s vulnerability, distracting oneself, avoiding self-reflective thoughts, and so forth (Pyszczynski et al., 1999). The “second line of defense,” called distal defenses, are characterized by attempts to achieve literal (i.e., religious) or symbolic death transcendence through support of one’s worldview or self-esteem (Hirschberger, 2010, p. 205). Symbolic death transcendence or symbolic immortality requires that some part of one’s self—one’s family, achievements, community—will continue to exist after death.

Understanding these two levels of reactions to mortality salience (avoidance and transcendence) can help to explain common issues in estate planning and planned giving. The “proximal” defense suggests that the most common initial reaction will be to avoid or postpone the topic. This matches the reality that, even among older adults, most have no will or trust (James et al., 2009). For fund-raisers, the enemy of the planned gift often isn’t “no”; the enemy is “later.” Combatting this may involve creating deadlines, making appointments, or promoting time-limited campaigns, all in an attempt to work against the inclination to avoid the topic altogether. For example, Rosen (2011) in his book on planned gift marketing pointed to the example of a challenge gift where a donor agreed to match 10% of bequests, up to US$10,000 per donor, signed prior to a deadline.

However, once problems of avoidance are overcome, the distal defenses to mortality-salience become more relevant. Symbolic immortality requires something, identified with the decedent, which will live beyond them. For most, the desire for symbolic immortality may be expressed by benefitting descendants and immediate family members who will live on beyond them. (It may also help to explain why the most significant demographic predictor of charitable estate planning is the absence of children; James, 2009a.) Among those who do leave charitable bequests, the desire for symbolic immortality is consistent with the inordinately large share of bequest dollars that go to permanent private foundations, typically bearing the deceased’s name (James, 2009b). It further suggests that charitable bequest donors will be particularly interested in lasting gifts (endowments, named buildings, scholarship funds, etc.) made to stable organizations.