Abstract
Life review (LR) is considered an important developmental process for promoting psychological and spiritual well-being and achieving ego integrity in older people. Some studies of LR with a general older adult population have revealed that LR interventions promoted ego integrity (Haight, Michel, & Hendrix, 2000), enhanced psychological well-being (Chiang, Lu, Chu, Chang, & Chou, 2008), improved personal meaning (Westerhof, Bohlmeijer, van Beljouw, & Pot, 2010), increased life satisfaction (Bohlmeijer, Roemer, Cuijpers, & Smit, 2007; Haight et al., 2000), increased self-esteem (Haight et al., 2000), and improved adaptation (Chiang et al., 2008).
LR has often been used with older adults who are diagnosed with clinical depression or are potentially at high risk of depression, such as those who are homebound or residents of nursing homes (e.g., Haight, Michel, & Hendrix, 1998) or individuals facing stressful events or transitions (Haight et al., 1998; Kennedy & Tanenbaum, 2000; Rosenblum & Corn, 2002). In addition, LR has been used in several different target populations, including demented older adults and non-demented older adults who are searching for meaningful engagement (Lin, Li, & Tabourne, 2011). Thus, depressive symptoms are often included as outcome measures in LR studies, whether or not the participants were clinically diagnosed.
A meta-analysis by Bohlmeijer, Smit, and Cuijpers (2003) included studies of the effectiveness of LR and reminiscence on depressive symptoms across different older adult target groups. For 19 controlled outcome studies, an overall effect size of 0.67 (95% confidence interval [CI] = [0.41, 0.92]) was found, indicating a statistically and clinically significant effect of reminiscence and LR on depressive symptomatology in older adults. The effect was greater in studies where participants had high depressive symptomatology (d = 1.23) than in studies where participants did not (d = 0.37). Also, effect size was greater for community-dwelling older adults than adults living in nursing homes or residential care. Moreover, a subsequent meta-analysis of 13 controlled outcome studies was conducted comparing the effects of LR or reminiscence on psychological well-being. The average effect size was moderate (d = 0.68; 95% CI [0.46, 0.87]). The effects of structured LR were found to be significantly higher (d = 1.04) than the effects of unstructured reminiscence (d = 0.40; Bohlmeijer et al., 2007).
A LR is typically structured around one or more life themes, most often family themes—ranging from one’s own childhood to the experience of being a parent or a grandparent—and work themes—ranging from first job through career to retirement. Other commonly used themes, although by no means the only ones, are as follows: major turning points; impact of major historical events; role of education, health, holidays, music, literature, or art in one’s life; experiences with aging, dying, and death; and meaning, values, and purpose. LR interventions are likely to explicitly involve a process of evaluation in which participants are asked to examine how their memories contribute to the meaning of their life, and they may then work at coming to terms with more difficult memories.
More specifically, our LR intervention was based on research about autobiographical memory (AM). AM is an aspect of memory that is related to the recollection of past events that a person has experienced (Williams et al., 2007). It is central to human functioning, as it is of fundamental significance for the individual’s sense of self and goal orientation (Conway & Pleydell-Pearce, 2000). Depressed adults have been shown to have difficulty retrieving specific AMs in comparison with non-depressed controls (for reviews, see Healy & Williams, 1999; Williams, 1996). Specific memories include sufficient detail to identify one particular incident (e.g., “the day I got married”). Reduced AM specificity (rAMS) is considered a cognitive marker and predictor of the course of depression (Sumner, Griffith, & Mineka, 2010). That is, individuals with depression, those in remission, and formerly depressed individuals all have significant difficulties providing specific memories (Williams et al., 2007). Research has shown that this phenomenon, also called overgeneral (autobiographical) memory (OGM), is more than just a symptom of depression; it functions as a vulnerability factor or trait marker (see Williams et al., 2007, for a more extensive discussion). Importantly, research suggests that rAMS can be modified (Williams, Teasdale, Segal, & Soulsby, 2000), and that improvements in rAMS are associated with reduced rumination and cognitive avoidance, and improved problem-solving skills (Raes, Williams, & Hermans, 2009).
The aim of this study was to examine the effects of autobiographical retrieval practice based on LR in Specific Positive Events in a community sample of older adults. This sort of intervention could be a useful tool to improve life satisfaction and well-being in older adults. We predicted that older adults who received practice would improve their subjective well-being, as reflected in decreased depressive symptoms, improvement in rewarding environmental experiences, and increased life satisfaction, in comparison with a control group composed of older adults who carried out their usual educational activities.
Method
Participants
The participants were all students of the “Aulas de Mayores” (specially designed courses for older adults) in the province of Alicante, which are administered by the Miguel Hernández University (UMH). These courses offer cultural, educational, and sporting activities aimed at promoting active aging. To calculate the sample size, estimated data of previous works have been used following the guidelines proposed by Friedman, Furberg, and DeMets (1998). Regarding the expected difference, Serrano, Latorre, Gatz, and Montañés (2004), who performed a similar study in older people with depressive symptomatology, found a reduction in the Center for Epidemiological Studies Depression Scale (CES-D) scores of 30%. With regard to the statistical distribution of the CES-D scores, results of previous works conducted with similar samples (Ros et al., 2011; Ros, Latorre, & Serrano, 2010) have been taken as reference (M = 9.60, SD = 5.91). Therefore, for an estimated mean equal to 10, we establish the minimum change in CES-D scores in 4 points (40%). For a risk value of 0.05 and an acceptable statistical power of 0.1, the required sample is 26 participants per group. Taking into consideration a replacement rate of 10%, a total of 60 volunteers were recruited and randomly assigned to one of two groups (experimental or control). It was checked that the groups were matched for age, gender, level of education, and cognitive and depression scores. Before the intervention started, one participant was hospitalized due to illness and so dropped out of the study (retention rate = 96.7%), and three participants from the control group opted not to participate (retention rate = 90%). There were no dropouts between pre-test and post-test.
The final sample included 55 older adults ranging in age from 53 to 89 (M = 65.35; SD = 8.45); 18 were women and 37 were men. With regard to educational levels, 40% of the participants had completed primary education, 40% had completed secondary, and 20% had attended university. Most of the participants (67%) lived with their families, 20% lived alone, whereas the rest lived in some sort of shared accommodation. Most of the participants had no physical illness. Only 27.3% had a chronic illness, the most common of which were type 2 diabetes, high blood pressure, and heart disease. None of the participants had diagnosis of mental illness at the beginning of the intervention (experimental or control groups). Table 1 shows the characteristics of the two groups, which, as can be seen, have the same socio-demographic profile and verbal fluency scores.
Background Data for the Participants in the Study.
Measures
Verbal fluency
The FAS test (Rami, Serradell, Bosch, Villar, & Molinuevo, 2007) was used to control the impact of cognitive function on the possible effectiveness of the LR. This task is a phonetic verbal fluency test measuring the number of words the subject can say in a minute beginning consecutively with F, A, and S. The overall score is obtained by adding up the number of words pronounced by the subject for each of the letters. This test measures the frontal cognitive functions and more specifically, the executive functions.
Center for Epidemiologic Studies-Depression
Depressive symptoms were measured by the CES-D (Radloff, 1977; translated into Spanish by Latorre & Montañés, 1997). The CES-D is a 20-item self-report scale developed to screen for depressive symptomology in the general population, with each item scored on a 4-point scale from 0 (little or no experience of the symptom over the past week) to 3 (nearly constant experience of the symptom). Total scores range from 0 to 60, with higher scores indicating more depressive symptoms.
Life satisfaction
Life satisfaction was measured using the Life Satisfaction Index A (LSIA; Adams, 1969). The original Life Satisfaction Index (Neugarten, Havighurst, & Tobin, 1961) consisted of 20 “agree” or “disagree” attitude items. Adams (1969) reduced the scale to 18 questions, using Wood’s scoring method (Wood, Wylie, & Sheafor, 1969). This method assigns two points for positive answers, no points for negative answers, and one for “don’t know” answers, providing a range of 0 to 36, with the highest scores indicating the greatest satisfaction. The scale was translated into Spanish by Stock, Okun, and Gómez (1994).
Environmental Reward Observation Scale (EROS)
The EROS (Armento & Hopko, 2007) is an instrument with 10 items developed as an efficient, reliable, and valid self-report measure of environmental reward. The items measure increased behavior and positive affect as a consequence of rewarding environmental experiences. Items on the EROS are answered using a 4-point Likert-type scale, ranging from 1 (strongly disagree) to 4 (strongly agree), with the total score representing a sum of the 10 items. Total scores ranged from 10 to 40. The version used here is the Spanish adaptation by Barraca and Pérez-Álvarez (2010).
Written Autobiographical Memory Test (AMT)
The Written AMT (Raes, Hermans, de Decker, Eelen, & Williams, 2003) was adapted for the Spanish population by Ricarte, Latorre, and Ros (2013). Participants were given a booklet with 15 pages. A cue word was printed on each page, and instructions were displayed on the first page. The first two cues were practice items. Instructions and cue words were also read by the test administrator, and further explanations were provided both at the end of the instructions and at the end of the practice session to ensure that the participants understood the task. Ten cue words were presented (five positive and five negative) with words of positive and negative emotional valence presented in alternating order. Participants were asked to write down a specific memory for each cue. The instructions stated that the specific memory should be of one particular occasion or event that happened on a particular day at least 1 week ago. Each recalled memory had to be different. These instructions are standard for the AMT procedure (see Williams, Barnhofer, Crane, & Beck, 2005). When the 60-s time limit for a cue was reached, the participants were instructed to turn to the next cue. Cue words were screened from the Brittlebank, Scott, Williams, and Ferrier (1993) study and were matched for frequency, imageability, and familiarity by valence (positive/negative) using a standardized lexical program (Davis & Perea, 2005). Recalled memories of events that lasted less than 24 hr were coded as specific (e.g., “my wedding day”). Recalled memories of events that lasted more than 24 hr were coded as extended (e.g., “when I was at school”). Memories referred to repeatedly occurring class of events generally stored in categories such as persons, places, or activities were coded as categorical (e.g., “every argument with my husband”). The names of persons, animals, or objects that did not mention a life event were considered semantic associations (e.g., “my father”). Finally, any response that did not belong to any of the above categories was coded as no memory. Two researchers who were independently blinded to the experimental and control groups rated the responses of all of the participants. Five categories were established according to memory type: specific memories (MSp), categorical memories (MCat), extended memories (MExt), semantic associations (MSA), and no recall (NM). An inter-rater agreement of 91% was obtained. To complete the analysis, the specific memories were coded according to the valence of the cue word. Consequently, the MSp category was subdivided into MSp+ and MSp−.
Procedure
Following recruitment, the experimental group received the LR intervention in individual sessions (ReViSEP; Serrano et al., 2004). The active control (AC) group comprised a “media workshop” in which participants were formed into small groups of four or five participants and received six weekly sessions from an interviewer. This was also carried out individually, with the interviewer asking each participant to speak independently. It was ensured that the participants in the control group did not take part in any other educational activity during the 8 weeks it took to complete the procedure. In the first week, participants gave consent, received an explanation of the study, provided basic demographic data, were administered the verbal fluency test, and completed the pre-test of self-report measures and the AMT. This first session was conducted individually for all participants. During the second to seventh weeks, LR was carried out with the experimental group and AC with the control group. AM post-testing and self-report measures took place in the 8th week after the intervention was completed. All sessions took place in the university facilities.
Experimental group (LR)
The intervention in the experimental group was individually administered on a weekly basis by 4th- and 5th-year psychology students. Participants were told that the study was investigating the effects of memory recall on mood and that the interviews were designed to evoke memories. During the period from the 2nd to 7th week, the LR was carried out with the experimental group. The LR consisted of autobiographical retrieval practice that entailed focusing on a particular life period each week: (S1) Early Childhood, Family, and Home; (S2) Later Childhood and Adolescence; (S3) Young Adulthood; (S4) Older Adulthood; (S5) Summary and Evaluation; (S6) Evaluation and Integration. For each of six sessions, 14 questions were prepared (based on Haight & Haight, 2007), designed to prompt specific memories. This intervention program is similar to the one used in earlier studies (Serrano et al., 2004). The difference in this case is that two further sessions were included so as to adapt the interview to the model of Erikson’s (1959) eight stage life cycle model, following the framework of Haight and Haight (2007), but focusing on specific AMs. Examples of questions include the following: “What is the most pleasant situation you remember from your childhood?”; “When you were a child did your mother or father ever do anything really surprising that made you feel happy? What?”; “During adolescence, what moment do you remember as most special because it was your first kiss or because you shared something special with someone you loved?”; “Tell me about a day as an adolescent when you did something unusual which you remember with happiness?”; “Tell me about your proudest moment at work. What happened that day?”; “When you were young, did you ever feel glad because someone close to you recovered from a serious illness?”; “Do you remember an experience of seeing something beautiful for the first time, like a work of art, a new-born baby, etc.? Could you describe this moment? Where were you? How did it make you feel?”; “If everything in your life were to happen again in exactly the same way, what moment would you most like to re-live?”; and “What do you consider to be the most important thing you have done in your life and that you remember with great joy?” Participants’ reported memories were recorded and codified according to memory type (specific, categorical, extended, and semantic association) and emotional valence (positive, negative, or neutral).
Control group (AC group)
The AC group was designed to be similar in content and format to the educational activities typically conducted with participants in the “Aulas de Mayores.” These participants frequently take part in group activities. During the intervention period and so as to avoid participants in the control group taking part in an activity which could affect this study, a group activity “media workshop” was designed as an AC for the variable. The participants were randomly assigned to small groups of five to six participants, specifically four groups of five and one of six. The sessions were conducted by 2nd- and 3rd-year journalism students. Six students took part, one per group. The first three sessions focused on the “news of the day,” which was summarized, analyzed, and discussed in the group context. Sessions 4 to 6 were conducted as if the participants were part of the editorial office of a newspaper, distributing the different sections (national and international news, sports, weather, etc.) with the aim of collecting information so as to produce their own newspaper in the final session. To control the effect of face to face interaction, the sessions were planned in such a way that each participant had to speak individually to the group facilitator.
Statistical Analysis
The statistical analysis was conducted using SPSS for Mac (v. 19.0.). An ANOVA for repeated measures was used to evaluate the effects of LR on the mood state variables. An ANCOVA for repeated measures was used to evaluate the effect of LR on AMT variables, taking the pre-test CED-S as covariate, because in earlier studies, a correlation was found between this score and specific AMs (Williams et al., 2007). The pre-test/post-test score change was calculated for all measures as was the value of the effect size for change (Cohen’s d; Cohen, 1988). The paired t test was used to analyze the change in each of the groups. The percentage of change was also calculated. This measures the size of the mean score change with respect to a previous mean score; so that “percentage change” = (Mpost − Mpre)/Mpre, where Mpost and Mpre are the average of the scores after and before applying the treatment, respectively. The “effect sizes” were also calculated. They constitute a standard measure of the change produced by experimental manipulation and thus offer comparable information on the size of this change. The d values were calculated to estimate the effect size produced intra-group, that is, the degree to which the treatment changed the scores of a group without taking into account the other groups. For the calculations of “effect size,” the software packages Sample Power 2.0 for Windows (Borenstein, Rothstein, & Cohen, 2001) and The Effect Size Generator for Windows: Version 2.3. (Devilly, 2004) were used.
Finally, to assess the relationship between change in mood and change in AMs, additional correlation and regression analysis were also calculated.
Results
Descriptive Analyses
Table 2 shows the measures and standard deviations for both groups (LR and control) at the two times of measurement (pre-test and post-test), of all the dependent variables used in the study. The first results shown are those of the AMT, separating the total number of specific, categorical, and extended memories, and the semantic associations. In subsequent analyses, we will focus only on specific and categorical memories. The second results shown are those of the three tests used: EROS, LSI, and CES-D.
Means for the Measures (Direct Scores) for the Two Groups in Pre-Test and Post-Test.
Note. The values are M ± SD. AMT = Autobiographical Memory Test; EROS = Environmental Reward Observation Scale; LSI = Life Satisfaction Index; CES-D = Center for Epidemiological Studies Depression Scale.
Regarding AMT scores, there were no significant differences between groups (LR vs. control) in pre-test for no recall, t (53) = 0.04, p = .971; and for specific, t (53) = −0.51, p = .615; categorical, t (53) = 0.44, p = .665; or extended, t (53) = −0.90, p = .375, memories. A significant difference was found in semantic associations variable, t (53) = 2.44, p = .018.
With regard to well-being measures, there were no significant differences between LR and control groups in pre-test for the CES-D Scale, t(53) = 1.11, p = .270, and Life Satisfaction Index, t(53) = −0.30, p = .760. Nevertheless, a difference was observed in the EROS test scores, t(53) = −2.11, p = .040.
Effect of LR on Mood State
Results for depressive symptomatology indicated a significant interaction between time and group, F(1.53) = 8.49, p = .005,
For life satisfaction (LSI), the main effect of repeated measures was found to be significant, F(1, 53) = 4.75, p = .034,
No significant effects were observed for the EROS test scores. The EROS test score increased by 5.8% in the experimental group whereas in the control group, it remained practically the same.
Effect of LR on Access to Specific AMs
Results of the 2 × 2 × 2 ANCOVA showed a second-order interaction of Time × Memory type × Group, F(1.52) = 4.87, p = .032,
To see whether the increase in the number of specific memories differs according to the valence of the key word, two differences were calculated: (a) the number of specific positive memories at post-test minus the number of specific positive memories at pre-test; and (b) the number of specific negative memories at post-test minus the number of specific negative memories at pre-test. Using as dependent variable the change in the specific positive memories, we observed that the LR group increased more than the control group, t(53) = 2.81, p = .007. No differences were observed in the case of the change in specific negative memories, t(53) = 0.32, p = .746. The number of specific positive memories in the LR group increased from a mean of 2.28 (SD = 1.75) at pre-test to a mean of 3.03 (SD = 1.59) at post-test. However, in the control group, it actually decreased slightly from a mean of 2.81 (SD = 1.68) at pre-test to a mean of 2.62 (SD = 1.60) at post-test.
Relation Between Emotional Variables and Retrieval
In the total sample (N = 55), the three measures of mood state are related at pre-test. Life satisfaction correlates positively with gratification (rxy = .52, p < .001) and negatively with depressive symptomatology (rxy = −.51, p < .001). Depressive symptomatology, in turn, correlates negatively with gratification (rxy = −.70, p < .001). At post-test, life satisfaction correlates negatively with the increase in specific negative memories (rxy = −.26, p = .058). The pre-test/post-test increase in life satisfaction correlates positively with the total number of specific memories at pre-test (rxy = .27, p = .046).
Additional Analyses
Additional analyses were conducted to see whether change in specific memories was related to change in the outcome measures. A change score for specific negative memories was calculated by taking the difference between the number of specific negative memories generated at post-test and the number generated at pre-test. A change score for specific positive memories was also calculated following the same procedure. Three regressions were carried out, using all 55 participants, with post-test scores on CES-D, EROS, and LSI as the outcomes and three predictors: pre-test scores on the corresponding measures and change in specific positive and specific negative memories. For all three outcomes, post-test scores were significantly predicted by pre-test scores (all ps < .000). With regard to change scores in specific memories, only change in specific negative memories was a significant predictor for life satisfaction (B = −1.26, p = .024, 95% CI [−2.35, −0.17]). This result suggests that participants with the greatest decrease in production of specific negative memories were also those who improved most in life satisfaction.
Discussion
The main aim of this study was to evaluate the effect of a six-session LR training therapy based on specific positive events on non-depressed older adults routinely taking part in active aging programs. We studied whether older adults who receive LR therapy improve their mood state and whether the improvement might be related to changes in access to specific memories. The findings support the idea that this technique is beneficial in the reduction of depressive symptomatology. Despite the fact that initially, the participants had few depressive symptoms, the LR therapy resulted in a significant decrease in symptoms relative to the control group. Although this therapy has already been tested with older adults suffering from clinical (Serrano et al., 2012) or subclinical depression (Serrano et al., 2004), extending its potential application to older people without depressive symptoms may bring many benefits. If we compare our data with those from the meta-analysis conducted by Bohlmeijer et al. (2007), we can conclude that the effect size we found for structured LR (d = 0.59) is smaller than that of people with high depressive symptomatology (d = 1.23) yet larger than that of those who have few symptoms (d = 0.37). As this is a technique that can be carried out by trained assistive personnel, it could be used as a simple complement to active aging programs in both residential and community environments.
Life satisfaction scores increased in both groups. These results lead us to consider that participation in any sort of training activity improves life satisfaction in older people. However, the effect of AM on life satisfaction must be emphasized because the participants with the largest increase in the number of specific memories were also those with the greatest increase in LSI. Previous studies have found an association between the capacity of access to specific memories and life satisfaction in older adults with or without depressive symptoms (Latorre et al., 2013; Ricarte et al., 2011; Ros & Latorre, 2010; Ros et al., 2010; Serrano, Latorre, & Gatz, 2007; Serrano et al., 2004). In their study, Ricarte et al. (2011) found a positive correlation between specific memory and LSI in the non-depressed group in comparison with the group of participants with depression. More recently, this type of association between specific memory and LSI has also been found in older adults with a low CES-D score in comparison with older adults with a high score who are not, however, clinically diagnosed as suffering from major depression (Latorre et al., 2013). In short, greater gains in access to specific memories are associated with a higher level of life satisfaction in older adults without depressive symptoms. Although a similar trend is observed to that in other studies (Serrano et al., 2004; Serrano et al., 2012), in our study, the effect is related to the decrease in access to specific negative memories. In short, this result is in line with the findings of Ros and Latorre (2010), who showed that older adults retrieved fewer negative memories than younger adults in response to negative cue words. Training in access to specific positive memories improves life satisfaction by reducing access to negative memories.
As predicted, the LR intervention generated an increase in the number of specific memories and a decrease in categorical memories, thus confirming the results of earlier studies with depressed older adults (Serrano et al., 2004; Serrano et al., 2012). However, in this case, it must be taken into account that the baseline level of specific memory was already high (M = 5.40, SD = 2.96) and similar to that found in a recent study (Ros et al., 2010) in which 46 older adults with a medium-high educational level participated (M = 5.63, SD = 1.88). In previous studies where this type of intervention was conducted with depressed older adults (Serrano et al., 2004; Serrano et al., 2012), the baseline level of specific AM was much lower in both participants with symptoms of depression (M = 3.15, SD = 3.10) and those with clinical depression (M = 2.40, SD = 2.20). The results of our study suggest that LR training is an effective tool for improving access to specific AMs in both depressed and non-depressed older adults.
The lack of effects of the treatment on the EROS test scores may be explained by the nature of the test. This test evaluates the perception of reward produced by our environment using items such as “many of my daily activities are pleasant” or “I am satisfied with the achievements in my life.” In this case, as the participants were all active people, frequently involved in gratifying activities, the intervention did not significantly increase the sensation of reward that the participants regularly find in their own environment. In fact, the baseline scores are higher than the values found in non-clinical population and much higher than those found in clinical population (Barraca & Pérez-Álvarez, 2010).
Limitations
It is important to acknowledge the limitations of this study. The first is that a follow-up of the participants was not conducted, and therefore, we cannot know whether the changes found are maintained over the passage of time. The second limitation lies with the evaluation instruments used. In this study, to compare the results with previous studies using the same technique (Serrano et al., 2004; Serrano et al., 2012), we chose to measure the results clinically. In subsequent studies along these lines, it would be advisable to use measures of life satisfaction and satisfactory aging, which are more sensitive to the changes in non-depressed older adults. The third limitation is related to the sample size. Although differences were found in the most important variables, a larger sample size would have meant an improvement in statistical power. Furthermore, these results cannot be generalized to populations of older people with a lower level of education. Future studies need to be focused on the efficacy of LR in more representative populations.
Implications for Therapeutic Interventions
In summary, we found that a six-session LR therapy focused on specific positive events decreases the depressive symptoms and increases access to specific AMs in non-depressed older adults. LR can therefore provide an excellent tool for people with mild psychological distress who need support in coping with transitions or adversities in life (Webster, Bohlmeijer, & Westerhof, 2010). However, it can also be used as a preventive strategy included in active aging programs. The fundamental mechanism in the use of this technique lies in the greater access to specific memories, which can provide an additional factor of protection. LR training can thus contribute to satisfactory active aging (Bowling, 1993, 2006, 2007; Bowling & Illiffe, 2011).
Footnotes
Acknowledgements
We would like to give our thanks to Margaret Gatz (University of Southern California) for her collaboration in the revision of this article.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Regional Government of Castilla-La Mancha [Consejería de Educación y Ciencia de Castilla-La Mancha, Grant PII1I09-0274-8863] and the Ministry of Science and Innovation [Ministerio de Ciencia e Innovación, Grant PSI2010-20088].