ArticlePDF Available

Extraneous Factors in Judicial Decisions

Authors:

Abstract and Figures

Are judicial rulings based solely on laws and facts? Legal formalism holds that judges apply legal reasons to the facts of a case in a rational, mechanical, and deliberative manner. In contrast, legal realists argue that the rational application of legal reasons does not sufficiently explain the decisions of judges and that psychological, political, and social factors influence judicial rulings. We test the common caricature of realism that justice is "what the judge ate for breakfast" in sequential parole decisions made by experienced judges. We record the judges' two daily food breaks, which result in segmenting the deliberations of the day into three distinct "decision sessions." We find that the percentage of favorable rulings drops gradually from ≈ 65% to nearly zero within each decision session and returns abruptly to ≈ 65% after a break. Our findings suggest that judicial rulings can be swayed by extraneous variables that should have no bearing on legal decisions.
Content may be subject to copyright.
Extraneous factors in judicial decisions
Shai Danziger
a,1
, Jonathan Levav
b,1,2
, and Liora Avnaim-Pesso
a
a
Department of Management, Ben Gurion University of the Negev, Beer Sheva 84105, Israel; and
b
Columbia Business School, Columbia University, New York,
NY 10027
Edited* by Daniel Kahneman, Princeton University, Princeton, NJ, and approved February 25, 2011 (received for review December 8, 2010)
Are judicial rulings based solely on laws and facts? Legal formalism
holds that judges apply legal reasons to the facts of a case in a ra-
tional, mechanical, and deliberative manner. In contrast, legal real-
ists argue that the rational application of legal reasons does not
sufciently explain the decisions of judges and that psychological,
political, and social factors inuence judicial rulings. We test the
common caricature of realism that justice is what the judge ate
for breakfastin sequential parole decisions made by experienced
judges. We record the judgestwo daily food breaks, which result in
segmenting the deliberations of the day into three distinct deci-
sion sessions.We nd that the percentage of favorable rulings
drops gradually from 65% to nearly zero within each decision
session and returns abruptly to 65% after a break. Our ndings
suggest that judicial rulings can be swayed by extraneous variables
that should have no bearing on legal decisions.
decisionmaking
|
legal realism
|
mental depletion
|
expert
decisionmaking
|
ego depletion
Does the outcome of legal cases depend solely on laws and
facts? Legal formalism holds that judges apply legal reasons to
the facts of a case in a rational, mechanical, and deliberative
manner (1, 2). An alternative view of the lawencapsulated in the
highly inuential 20th century legal realist movementis rooted in
the observation of US Supreme Court Justice Oliver Wendell
Holmes that the life of the law has not been logic; it has been
experience(3). Realists argue that the rational application of
legal reasons does not sufciently explain judicial decisions and
that psychological, political, and social factors inuence rulings as
well (4). The realist view is commonly caricaturized by the trope
that justice is what the judge ate for breakfast(5). We empirically
test this caricature in the context of sequences of parole decisions
made by experienced judges (mean experience = 22.5 y, SD = 2.5)
and, in so doing, demonstrate how extraneous factors can sway
highly consequential decisions of expert decision makers.
Prior research suggests that making repeated judgments or deci-
sions depletes individualsexecutive function and mental resources
(6), which can, in turn, inuence their subsequent decisions. For
instance, sequential choices between consumer goods can lead to an
increase in intuitive decisionmaking (7) as well as a reduced toler-
ance for pain in a subsequent task (8). Sequential choices and the
apparent mental depletion that they evoke also increase peoples
tendency to simplify decisions by accepting the status quo. German
car buyers, for instance, were more likely to accept the default at-
tribute level offered by a manufacturer later in a sequence of attri-
bute decisions than earlier, particularly when these choices followed
decisions between many alternatives that had required more mental
resources to evaluate (9). These studies hint that making repeated
rulings can increase the likelihood of judges to simplify their deci-
sions. We speculate that as judges advance through the sequence of
cases (whose order appears to be exogenously determined; see below
for a detailed discussion), they will be more likely to accept the de-
fault, status quo outcome: deny a prisonersrequest.
Materials and Methods
Our data consist of 1,112 judicial rulings, collected over 50 d in a 10-mo
period, by eight Jewish-Israeli judges (two females) who preside over two
different parole boards that serve four major prisons in Israel. Our prisoner
sample consisted of 727 Jewish-Israeli males (65.3%), 326 Arab-Israeli males
(29.3%), 50 Jewish-Israeli females (4.5%), and 9 Arab-Israeli females (0.9%).
The two parole boards process 40% of all parole requests in the country.
The prisons house felons convicted of crimes such as embezzlement, assault,
theft, murder, and rape. Each parole board is composed of one judge, as
well as a criminologist and a social worker who provide the judge with
professional advice. For each day we obtained the entire set of rulings. The
majority of the decisions in our sample (78.2%) consist of parole requests;
the remainder consist of parolee requests to change the terms of their pa-
role (e.g., a request to remove a tracking device) or requests by parole
candidates to change the terms of their incarceration (e.g., a request for
prison relocation). Our database includes the legal variables that appear in
the case le: number of previous incarcerations, gravity of crime committed,
months served, and whether a rehabilitation program would be available
should the prisoner be granted parole (98.3% of prisoners had such a pro-
gram in place). [A judge with 40 years of experience on the bench, two
criminal attorneys, and two prison wardens with 10 years experience serving
on the parole board, independently ordered the gravity of offense for the 7
classes of crimes committed. Ordering was identical for the ve experts, and
ranged from misdemeanor (1) to felony (7).] The judge was not provided
these details in advance; the information was provided by a clerk only when
the prisoner (and his or her attorney) appeared before the parole board.
Every day a judge considered 1435 cases (see SI Materials and Methods, S1
for details) in succession (M= 22.58, SD = 4.67), and each case deliberation
lasted 6 min (M= 5.98, SD = 5.13, Max = 40.00). Our data include the time
of day in which the prisoners request was considered and its ordinal posi-
tion in the sequence of decisions for that day.
Executive function can be restored and mental fatigue overcome, in part, by
interventions such as viewing scenes of nature (10), short rest (11), experiencing
positive mood (12), and increasing glucose levels in the body (ref. 13; for
a review see ref. 14). In our data, we record the two daily food breaks that the
judge takesa late morning snack and lunchwhich serve to break up the
days deliberations into three distinct decision sessions.Such a break may
replenish mental resources by providing rest, improving mood, or by in-
creasing glucose levels in the body. The meal is typically served to the judge at
the bench and its timing, which is determined by the judge, varies by day. In
our sample, the start time of the morning food break ranged between 9:49
and 10:27 AM (snack consisting of a sandwich and fruit) and lasted an average
of 38.48 min (SD = 20.50, min = 6, max = 106); the start time of the afternoon
(lunch) break ranged between 12:46 and 2:10 PM and lasted an average of
57.37 min (SD = 22.00, min = 15, max = 110). The breaks were taken after an
average of 7.8 cases (SD = 4.51, min = 2, max = 28) in the morning session and
11.4 cases (SD = 5.14, min = 2, max = 25) in the postsnack/prelunch session.
Thus, our data enable us to test the effect of the ordinal position of a case on
the judges decision and the effect of the judge having taken a break to eat.
The judgesdecisions are classied into two categories, accept request
and reject request.Under the reject category, we include both nal
rejections as well as rejections that include a stipulation for review at a later
date (such delay decisions constitute 48.4% of the reject category). On av-
erage, such reviews occur 1 mo after the initial parole board review. Thus,
a decision to delay effectively maintains the status quo for the prisoner.
Overall, 64.2% of prisoner requests in our sample were rejected.
Author contributions: S.D., J.L., and L.A.-P. designed research; S.D., J.L., and L.A.-P. per-
formed research; J.L. analyzed data; and S.D. and J.L. wrote the paper.
The authors declare no conict of interest.
*This Direct Submission article had a prearranged editor.
1
S.D. and J.L. contributed equally to this work.
2
To whom correspondence should be addressed. E-mail: jl2351@columbia.edu.
This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.
1073/pnas.1018033108/-/DCSupplemental.
www.pnas.org/cgi/doi/10.1073/pnas.1018033108 PNAS Early Edition
|
1of4
SOCIAL SCIENCES
Results
We nd that the likelihood of a favorable ruling is greater at the
very beginning of the work day or after a food break than later in
the sequence of cases. This pattern is readily evident in Fig. 1,
which plots the proportion of favorable rulings by ordinal position
for 95% of the observations in each decision session. The plot
shows that the likelihood of a ruling in favor of a prisoner spikes at
the beginning of each sessionthe probability of a favorable
ruling steadily declines from 0.65 to nearly zero and jumps back
up to 0.65 after a break for a meal. Fig. 2 Aand Bpresents
a histogram of the probability of a favorable ruling for cases of
similar legal characteristics that appeared in one of the three
ordinal positions at the beginning versus at the end of a decision
session; from the perspective of the prisoner, there is a clear
advantage to appearing at the beginning of the session (i.e., either
at the beginning of the day or immediately following the break).
To account for the possible role of covariates in the patterns
depicted in Figs. 1 and 2, we used a logistic regression with rulings
as the dependent variable and a judge-specicxed effect to
control for the idiosyncratic tendencies of each judge (Table 1).
The key predictors were several different indicators of a cases
ordinal position: (i) dummy variables indicating the rst three
cases in a session, included to examine how judgments immediately
after a break differ from those that preceded or succeeded them;
(ii) dummies indicating in which of the three daily sessions the case
had appeared; and (iii) two types of ordinal position counters (one
indicating the ordinal position within the session and the other
indicating the ordinal position within the day, each used in a dif-
ferent regression specication). The covariates included all of the
legal attributes of the case that were available in the case le (se-
verity of crime, months served, previous incarcerations, and re-
habilitation program), prisoner demographics (sex, nationality),
and the proportion of favorable rulings to that point in the day. The
purpose of the latter was to control for the possibility that the
judges have a daily quotaof favorable decisions that they expect
to render, which, once lled, are followed by unfavorable decisions.
The positive sign and signicance of the dummy variables in-
dicating the rst three cases in each session conrms that the
pattern in Fig. 1 holds even while controlling for the legal
attributes of the case and for the overall tendency of the judges to
rule against the prisoner as the number of cases before them
mounts (i.e., the main effect of making repeated decisions). The
results are nearly identical when we restrict our analysis only to
parole requests (Table S1) and in analyses where we drop the two
most frequently occurring judges (Table S2) and each of the
judges in our sample (Tables S3S10). In addition, a plot similar
to Fig. 2 for each judge shows that every judge in our sample was
more likely to rule in favor of a prisoner at the beginning of
a session than at the end of a session (Fig. S1). Nested model tests
indicate that adding the ordinal position variables leads to better
model t(Table S11). Therefore, although our data do not allow
us to test directly whether justice is what the judge had for
breakfast, they do suggest that judicial decisions can be inuenced
by whether the judge took a break to eat.
We conducted an additional analysis to test the statistical ro-
bustness of the linear trend that is apparent between breaks in Fig.
1; regardless of the ordinal position counter we used, the trend
was signicant and negative (Table S12). We also conducted an
analysis using cumulative minutes elapsed in a session in lieu of
the ordinal position dummies as a predictor, as well as our control
variables. Cumulative minutes serve as a proxy for mental fatigue
among the judges. Similar to the results presented in Table 1, this
analysis shows that as cumulative time within a session increases,
the likelihood of a favorable ruling decreases (Table S13 and Fig.
S2). However, note that in an analysis that included both the
cumulative minutes variable and the ordinal position counter,
only the latter was signicant (Table S14). This analysis hints that
the apparent depletion exhibited by the judges is due to the act of
making decisions rather than simply elapsed time (this in-
terpretation should be viewed in light of the high correlation
between cumulative minutes and ordinal position, r= 0.72, P<
0.0001). Two indicators support our view that rejecting requests is
an easier decisionand, thus, a more likely outcomewhen
judges are mentally depleted: (i) favorable rulings took signi-
cantly longer (M= 7.37 min, SD = 5.11) than unfavorable rulings
(M= 5.21, SD = 4.97), t= 6.86, P<0.01, and (ii) written verdicts
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
Proporon favorable decisions
Ordinal posion
Fig. 1. Proportion of rulings in favor of the prisoners by ordinal position.
Circled points indicate the rst decision in each of the three decision ses-
sions; tick marks on xaxis denote every third case; dotted line denotes food
break. Because unequal session lengths resulted in a low number of cases for
some of the later ordinal positions, the graph is based on the rst 95% of the
data from each session.
*** 0.61 *** 0.61
*0.52+
*** 0.63
0.27
0.22 0.22+
0.09
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
4-12 13-24 25-36 >37
Proporon favorable decisions
Months served
First 3 Last 3
** * 0.50 *** 0.49
*0.40+
0.33+
0.10
0.20
0.12 +
0.21 +
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
4-12 13-24 25-36 >37
Proporon favorable decisions
Months served
First 3 Last 3
A
B
Fig. 2. Proportion of favorable decisions for male felons with a rehabilitation
program as a function of ordinal position, months served, and previous incar-
cerations. These histograms reect the rst three versus the last threedecisions
collapsed over the three decisions sessions. They are for illustrative purposes
and are based on a subsample of the data.Plus signs (+) indicate cellsizes of <20.
(A) Data for prisoners with no previous incarcerations. (B) Data for prisoners
with one previous incarceration. Asterisks indicate results of a difference be-
tween proportions test. *P<0.1, **P<0.05, ***P<0.01.
2of4
|
www.pnas.org/cgi/doi/10.1073/pnas.1018033108 Danziger et al.
of favorable rulings were signicantly longer (M= 89.61 words,
SD = 65.46) than written verdicts of unfavorable rulings (M=
47.36 words, SD = 43.99), t= 12.82, P<0.01.
Of the legally relevant control variables entered in the regres-
sions, only the prior number of incarcerations of the prisoner and
the presence of a rehabilitation program consistently exerted a
statistically signicant inuence on the judgesrulings. Prisoners
who displayed a tendency toward recidivism were less likely to
receive favorable judgments, as were prisoners who lacked a
planned rehabilitation program. The severity of the prisoners
crime and prison time served tended not to exert an effect on
rulings, nor did sex and ethnicity. The lack of a signicant effect of
prisoner ethnicity indicates that the Jewish-Israeli judges in our
sample treated prisoners equally regardless of ethnicity. Although
previous research does hint at the presence of effects of prisoners
and judgesrace on sentencing decisions, in some cases, as in ours,
such effects are weak or absent (1518).
A key aspect for interpreting the association between the or-
dinal position of a case and parole decisions is whether an un-
observed factor determines case order in such a way that yields the
pattern of results we obtain. For instance, if prisoners without
a rehabilitation program or recidivists were somehow more likely
Table 1. Results of analysis using dummies for the rst three decisions in a session
Specication
Variable 1 2 3 4
Overall decision count 0.078*** (0.020) 0.080*** (0.021)
Overall count including nondecisions 0.111*** (0.018) 0.111*** (0.019)
Session 1/decision 1 0.850** (0.377) 0.670* (0.370) ——
Session 1/decision 2 1.366*** (0.383) 1.236*** (0.381) 1.409*** (0.387) 1.268*** (0.383)
Session 1/decision 3 0.374 (0.351) 0.270 (0.351) 0.336 (0.354) 0.261 (0.353)
Session 2/decision 1 1.055*** (0.355) 0.789** (0.359) 1.064*** (0.358) 0.809** (0.362)
Session 2/decision 2 0.259 (0.337) 0.042 (0.341) 0.221 (0.339) 0.026 (0.343)
Session 2/decision 3 0.761** (0.337) 0.592* (0.339) 0.735** (0.339) 0.583* (0.340)
Session 3/decision 1 2.873*** (0.425) 2.677*** (0.431) 2.805*** (0.425) 2.642*** (0.431)
Session 3/decision 2 0.888** (0.453) 0.677 (0.460) 0.818* (0.456) 0.644 (0.462)
Session 3/decision 3 0.340 (0.660) 0.520 (0.666) 0.410 (0.662) 0.555 (0.667)
Session 1 0.341 (0.247) 0.788*** (0.263) 0.478* (0.253) 0.874*** (0.265)
Session 3 1.064*** (0.321) 0.608* (0.334) 0.943*** (0.326) 0.542 (0.338)
Severity of offense 0.051 (0.096) 0.068 (0.097) 0.018 (0.099) 0.039 (0.101)
Previous imprisonments 0.241*** (0.059) 0.234*** (0.059) 0.228*** (0.061) 0.222*** (0.062)
Months served 0.004 (0.003) 0.004 (0.003) 0.004 (0.003) 0.004 (0.003)
Rehabilitation program 2.465*** (0.809) 2.415*** (0.825) 1.974** (0.845) 1.907** (0.862)
Ethnicity (0 = Jew, 1 = Arab) 0.204 (0.156) 0.227 (0.157) 0.177 (0.160) 0.198 (0.161)
Sex (0 = male, 1 = female) 0.201 (0.299) 0.218 (0.301) 0.158 (0.305) 0.172 (0.307)
Proportion favorable decisions ——0.937*** (0.333) 0.631* (0.339)
2 Log likelihood 1135.215 1110.609 1067.232 1045.706
This table presents various xed effects logistic regression specications. The session x/decision y parameters are dummy variables
that indicate the rst three decisions in each of the three sessions. Note that in specications 3 and 4 there is no value for the very rst
decision of the day because the regression includes a term for proportion of favorable decisions, which requires there to have been at
least one other decision that day. Ethnicity and sex are dummy variables. SEs appear in parentheses. *P<0.10, **P<0.05, ***P<0.01.
0
1
2
3
4
Gravity of offense
Ordinal posion
0
1
2
3
4
Previous incarceraons
Ordinal posion
0
10
20
30
40
50
60
Months served
Ordinal posion
0
0.2
0.4
0.6
0.8
1
1.2
Proporon rehab
program
Ordinal posion
AB
CD
Fig. 3. Mean level of control variables by ordinal position.
Circled points indicate the rst decision in each of the three
sessions; tick marks on xaxis denote every third case; dotted
lines denote food break. (A) Data for gravity of offense. (B)
Data for previous incarcerations. (C) Data for months served.
(D) Data reecting the proportion of prisoners with a re-
habilitation program. Because unequal session lengths resul-
ted in a low number of cases for some of the later ordinal
positions, the graphs are based on the rst 95% of the data
from each session.
Danziger et al. PNAS Early Edition
|
3of4
SOCIAL SCIENCES
to appear before a food break, we would naturally nd a greater
proportion of rejections occurring before the food break as well.
A number of procedural factors preclude this possibility.
First and most critically, the judge both determines when the
break will occur during the course of the days proceedings and is
unaware of the details of the upcoming cases. Thus, the judge
cannot decide when to take a break based on information related
to the nature of the upcoming cases. So, in the example above,
a judge cannot decide to take a break because he or she knows that
prisoners after the break will have no previous incarceration re-
cord. Relatedly, the type of case (e.g., severity of the crime) that
the judge had just ruled on exerted no signicant effect on the
likelihood of taking a break (Table S15). Furthermore, the large
variability in break start times and durations attests to the fact that
their occurrence would be nearly impossible to predict by any of
the prison staff involved in the parole proceedings.
Second, the ordinal position of cases is, with rare exception,
determined by the arrival time of the prisoners attorney. The
attorneys are sequestered in a room where they are unable to view
the proceedings of the board and, therefore, are unaware of any of
the rulings of the judge, how many prisoners preceded their cli-
ents case, or when and whether the food break occurred (after
the boards deliberations, attorneys exit through a different door).
Thus, by design they cannot learn about the advantage of
appearing after a break. Indeed, a survey administered to a sam-
ple of these attorneys after the primary data collection period
indicated that they were unaware of the effect of ordinal position
on rulings (see SI Materials and Methods, S2 for details). A similar
survey administered to parole board members (judges, criminol-
ogists, and social workers) revealed the same results (see SI
Materials and Methods, S3 for details).
Because of the factors discussed above, we did not expect sig-
nicant correlations between ordinal position within either the day
or the session and the control variables in our data (SI Materials
and Methods, S4 and Table S16). Consistent with our expectations,
there does not appear to be a deliberate ordering based on the
characteristics of the prisoners (Fig. 3 ADand SI Materials and
Methods, S4); certainly there appears to be no effect of a food
break on the type of prisoner appearing before the judge. Note that
although there was a slight but signicant correlation between
recidivism and ordinal position in the day, this correlation was not
signicant within a decision session, i.e., between breaks. Thus, it
cannot explain the spikes in favorable decisions after breaks.
Another factor that can plausibly explain our effect is that
judges might have a certain proportion of decisions that they expect
to be favorable, and once thisquotais lled, then unfavorable de-
cisions follow. As we explain earlier, we tested this possibility em-
pirically by including a variable that computed the proportion of
favorable decisions up to thatpoint in the day (Table1, specications
3 and 4). Regardless of the analysis we conducted, the parameter
estimate was positive and signicant, suggesting that a judge who
made a large proportion of favorable rulings up to a certain point
was, in fact, more likely to rule favorably in a subsequent case.
Discussion
We have presented evidence suggesting that when judges make
repeated rulings, they show an increased tendency to rule in favor of
the status quo. This tendency can be overcome by taking a break to
eat a meal, consistent with previous research demonstrating the
effects of a short rest, positive mood, and glucose on mental re-
source replenishment (1113). However, we cannot unequivocally
determine whether simply resting or eating restores the judges
mental resources because each of the breaks was taken for the
purpose of eating a meal. We also cannot ascertain whether taking
a break improved the judgesmood because mood was not mea-
sured in our study. Furthermore, although we interpret our ndings
through the lens of mental depletion, we do not have a direct
measure of the judgesmental resources and, thus, cannot assess
whether these change over time. Nevertheless, our results do in-
dicate that extraneous variables can inuence judicial decisions,
which bolsters the growing body of evidence that points to the sus-
ceptibility of experienced judges to psychological biases (19, 20; for
a review, see ref. 21). Finally, our ndings support the view that the
law is indeterminate by showing that legally irrelevant situational
determinantsin this case, merely taking a food breakmay lead
a judge to rule differently in cases with similar legal characteristics.
Although our focus has been on expert legal decisions, we sus-
pect the presence of other forms of decision simplication strate-
gies for experts in other important sequential decisions or judg-
ments, such as legislative decisions, medical decisions, nancial
decisions, and university admissions decisions. Our ndings add to
the literature that documents how experts are not immune to the
inuence of extraneous irrelevant information (2224). Indeed, the
caricature that justice is what the judge ate for breakfast might be
an appropriate caricature for human decisionmaking in general.
ACKNOWLEDGMENTS. We thank Jim Bettman, Brett Gordon, Michael Heller,
Eric Johnson, Daniel Kahneman, Itzhak Levav, Orly Lobel, Oded Netzer, Jeff
Rachlinski, Derek Rucker, Uri Simonsohn, Richard Thaler, and Andrew
Wistrich for comments.
1. Leiter B (2005) The Blackwell Guide to Philosophy of Law and Legal Theory, eds
Edmundson W, Golding M (Blackwell, Oxford), pp 5066.
2. Neuborne B (1992) Of sausage factories and syllogism machines: Formalism, realism
and exclusionary selection techniques. NYU L Rev. 67:419449.
3. Holmes OW (1881) The Common Law (Little, Brown, Boston).
4. Frank J (1930) Law and the Modern Mind (Brentanos, New York).
5. Kozinski A (1993) What I ate for breakfast and other mysteries of judicial decision
making. Loyola LA L Rev 26:993.
6. Muraven M, Baumeister RF (2000) Self-regulation and depletion of limited resources:
Does self-control resemble a muscle? Psychol Bull 126:247259.
7. Pocheptsova A, Amir O, Dhar R, Baumeister RF (2009) Deciding without resources:
Resource depletion and choice in context. J Mkt Res 46:344355.
8. Vohs KD, et al. (2008) Making choices impairs subsequent self-control: A limited-
resource account of decision making, self-regulation, and active initiative. J Pers Soc
Psychol 94:883898.
9. Levav J, Heitmann H, Herrmann A, Iyengar SS (2010) Order in product customization
decisions: Evidence from eld experiments. J Polit Econ 118:274299.
10. Kaplan R, Kaplan S (1989) The Experience of Nature: A Psychological Perspective
(Cambridge Univ Press, New York).
11. Tyler JM, Burns KC (2008) After depletion: The replenishment of the selfs regulatory
resources. Self Ident 7:305321.
12. Tice DM, Baumeister RF,Shmueli D, Muraven M (2007)Restoring the self: Positiveaffect
helps improve self-regulation following ego depletion. J Exp Soc Psychol 43:379384.
13. Gailliot MT, Baumeister RF (2007) The physiology of willpower: Linking blood glucose
to self-control. Pers Soc Psychol Rev 11:303327.
14. Hagger MS, Wood C, Stiff C, Chatzisarantis NLD (2010) Ego depletion and the
strength model of self-control: a meta-analysis. Psychol Bull 136:495525.
15. Steffensmeier D, Britt CL (2001) Judgesrace and judicial decision making: Do black
judges sentence differently. Soc Sci Q 82:749764.
16. Gazal-Eyal O, Sulitzeanu-Kenan R (2010) Let my people go: Ethnic in-group bias in
judicial decisions-Evidence from a randomized natural experiment. Journal of Empir
Leg Stud 7:403428.
17. Blair IV, Judd CM, Chapleau KM (2004) The inuence of Afrocentric facial features in
criminal sentencing. Psychol Sci 15:674679.
18. Spohn C.C (2000) Criminal Justice 2000: Vol. 3. Policies, Processes, and Decisions of
the Criminal Justice System, ed Horney J (US Department of Justice, Washington, DC),
pp 427501.
19. Guthrie C, Rachlinski JJ, Wistrich AJ (2001) Inside the judicial mind. Cornell Law Rev
86:777830.
20. Guthrie C, Rachlinski JJ, Wistrich AJ (2007) Blinking on the bench: How judges decide
cases. Cornell Law Rev 93:144.
21. Vidmar N (2011) The psychology of trial judging. Curr Dir Psychol Sci 20:5862.
22. Northcraft GB, Neale MA (1987) Experts, amateurs, and real estate: An anchoring-
and-adjustment perspective on property pricing decisions. Organ Behav Hum Decis
Process 39:228241.
23. Englich B, Mussweiler T, Strack F( 2006)Playing dice with criminal sentences: The inuence
of irrelevant anchors on expertsjudicial decision making. Pers Soc Psychol Bull 32:
188200.
24. Dhami MK (2003) Psychological models of professional decision making. Psychol Sci
14:175180.
4of4
|
www.pnas.org/cgi/doi/10.1073/pnas.1018033108 Danziger et al.
Supporting Information
Danziger et al. 10.1073/pnas.1018033108
SI Materials and Methods
S1. Number of Cases Viewed by the Judge. The total number of cases
thejudgeviewed in a day wasgreater than what we reportinthe main
text (M= 27.86, SD = 4.43). The reason for the discrepancy is that
some cases are brought before the judge after the state prosecutor
and defense attorney had already come to a resolution regarding
the prisoners request and were simply presenting the agreement
to the judge for nal approval. In our sample, the judge approved
every agreement. This situation is almost universally the case also
outside our sample. Thus, we omit these cases because they do not
represent actual decisions by the judge about which party to favor
and, hence, cannot be used in our empirical test.
S2. Survey of Attorneys. To ascertain whether the criminal defense
attorneys who represent the prison population in our study were
aware of a possible effect of order on judicial decisions, we dis-
tributed a survey among 23 lawyers of varying levels of experience
(M= 9.8 y experience, SD = 8.3). They were asked to indicate
the factors they thought inuence a decision by a judge to grant/
deny parole. The four most frequently mentioned factors were
having a rehabilitation program in place (n= 22, 95.6%), the
number of previous incarcerations (n= 17, 73.9%), severity of the
offense (n= 13, 56.5%), and prisoner behavior during the in-
carceration (n= 10, 43.4%). No other factor was mentioned by
more than 6 lawyers (26%). None of the lawyers mentioned or-
dinal position.
Next, on a 17 scale ranging from not at all(1) to to a large
degree(7), the lawyers were asked to rate the extent to which se-
verity of offense, prisoner ethnicity, prisoner sex, months in-
carcerated, number of previous incarcerations, having a rehabili-
tation program in place, the ordinal position of the case in the case
sequence, prisoner age, prisoner health, sex of the judge, and the
marital status of the prisoner inuences the decision of the judge to
grant/deny parole. Number of previous incarcerations (M= 6.13,
SD = 0.91), having an approved rehabilitation program (M=6.0,
SD = 1.17), severity of offense (M= 5.65, SD = 1.33), and months
in prison (M= 5.21, SD = 1.34) were rated as the most inuential
factors. The least inuential factors were ordinal case position
(M= 2.78, SD = 1.44) and prisoner ethnicity (M= 2.65, SD =
1.36). The importance score for ordinal case position was signi-
cantly lower (all Pvalues<0.0001) than that of each of the four most
inuential factors.
In summary, using two different measures, we do not nd
evidence to suggest that a sample of the lawyers present at the
parole hearings are aware of the strong effect that ordinal case
position can have on rulings.
S3. Survey of Parole Board Members. In addition to surveying the
lawyers, we also investigated whether members of the parole board
were aware of the effect of order. Sixteen parole board members
were asked to rate the extent to which severity of offense, prisoner
ethnicity, prisoner sex, months incarcerated, number of previous
incarcerations,having a rehabilitation program in place, the ordinal
position of the case in the case sequence, prisoner age, prisoner
health, the quality of the lawyer, the mood of the judge, prison
location, the seasonin the year and the marital status of the prisoner
inuences the decision of the judges to grant/deny parole. Four-
teeen members completed this part of the survey. Having an ap-
proved rehabilitation program (M= 6.50, SD = 0.65), number of
previous incarcerations (M= 5.85, SD = 0.86), severity of offense
(M= 5.14, SD = 1.23), and months in prison (M= 4.07, SD =
1.49) were rated as the most inuential factors. The least in-
uential factors were the prison location (M= 1.53, SD = 1.66),
prisoner ethnicity (M= 1.57, SD = 0.93), the season in the year
(M= 2.00, SD = 1.68), ordinal case position (M= 2.00, SD =
1.10), the mood of the judge (M= 2.07, SD = 1.03) and prisoner
sex (M= 2.28, SD = 1.48). The importance score for ordinal case
position was signicantly lower (all Pvalues <0.001) than that of
each of the four most inuential factors. Furthermore, the im-
portance score of ordinal position did not differ from any of the
unimportant factors (all Pvalues >0.27).
Next, we presented the 16 parole board members with three
written descriptions of possible relations between the decision of
the judge to grant/deny parole and ordinal case position. We
asked them to select the one that they believed best represented
the judges decisions as a function of ordinal case position. The
rst description was one in which there is no relation between
the order of cases and the judges decision to grant/deny parole.
The second description was one in which the probability of re-
lease increased from the rst case to the last case in each of the
three decision sessions in the day. The third description was one
that matched the pattern we nd in our data. None of the parole
board members indicated that the third description t the de-
cision pattern of the judges. Fifteen of 16 indicated that there
was no relation between ordinal case position and decisions, and
one member indicated that the second description t the de-
cision pattern of the judges.
In summary, using two different measures, we do not nd
evidence to suggest that a sample of the parole board members
are aware of the strong effect that ordinal case position can have
on rulings. Although this lack of awareness might seem surprising
at rst blush, it is worth noting that, even though the drop in
prisoner releases is dramatic, in most cases it is not quite as
dramatic as presented in Fig. 1. For instance, if one were to
examine only 80% of the cases before the judge in each decision
session, the drop in probability of release is around 45% rather
than 65%, as is evident when one plots 95% of the cases (i.e., Fig.
1). This drop is precipitous, but perhaps the fact that in most cases
the probability does not drop to zero reduces the likelihood that
the judges will perceive the presence of an order effect.
S4. Correlations Between Ordinal Position Indicators and Variables
Reecting Case Severity. We tested whether order of cases was
random by calculating the correlation between various indicators
of ordinal position in a decision session and variables that reect
the severity of the prisonerscrimes (severity of crime, months in
prison) as well as his or her history of recidivism (previous in-
carcerations). The ordinal position measures that we used in our
analysis were as follows: (i) a simple counter that increased for
each decision in the session; (ii) a counter that corresponded to
the overall number of cases brought before the judge within
the session, including those that had been agreed upon between
the prosecution and defense (SI Materials and Methods, S1); (iii)
the cumulative number of minutes spent deliberating within the
session up to that case; (iv) a counter that corresponded to the
overall number of decisions made in the day; and (v) an indicator
of which of the daily sessions the case appeared in. Table S16
presents the 20 correlations that we calculated; in the vast ma-
jority of cases, they are not statistically signicant. There was a
very mild negative correlation between two of the ordinal position
measures and severity of crime, such that prisoners convicted of
more severe crimes were slightly more likely to appear before the
judge earlier in each session. Note that this correlation predicts
that rulings early in the sequence would be less likely to favor the
Danziger et al. www.pnas.org/cgi/content/short/1018033108 1of10
felon, which is the opposite of what we nd. A very mild positive
correlation emerged between two other ordinal position mea-
sures and recidivism, indicating that recidivists were slightly more
likely to appear later in the day. In this regard, several points are
noteworthy. First, the correlations with recidivism are small and
are only signicant for the overall day ordinal position counter but
not the session ordinal position counter; thus, they cannot explain
the spikes in favorable decisions after a break. Second, note that
the 4 (of 20) signicant correlations are small and in opposite
directions of each otherthey are not consistent and, thus, do not
appear to indicate a pattern of systematic ordering of the cases
that might be giving rise to our ndings.
In a related analysis we examined the mean level of prisoner
characteristics for the three prisoners that appeared before and
after each of the two daily breaks. This analysis leads to similar
conclusions as the ones indicated by the correlations. There were
no signicant differences between the rst three cases and the last
three cases in a session with regard to the percentage of prisoners
with a rehabilitation plan (P= 0.675; rst three: 98.1%; last three:
98.5%), months of incarceration (P= 0.24; rst three: M= 31.43;
SD = 40.41; last three: M= 28.2; SD = 31.99), and number of
previous incarcerations (P= 0.695; rst three: M= 1.91; SD =
1.53; last three: M= 1.95; SD = 1.51). Finally, crime severity was
higher (P= 0.04) in the rst three cases (M= 2.92; SD = 1.03)
than the last three cases in a session (M= 2.77; SD = 0.91),
a pattern that predicts the opposite of the effect we nd because
presumably crime severity should decrease likelihood of release.
Fig. S1. The proportion of favorable decisions as a function of judge and ordinal position within a session. The data points reect proportions for the rst
three versus last three decisions in each of the three sessions, for each judge. On average a data point reects 16.00 decisions (min = 3, max = 27, SD = 7.12).
Danziger et al. www.pnas.org/cgi/content/short/1018033108 2of10
Fig. S2. Proportion of rulings in favor of prisoners by cumulative minutes of deliberation. Circled points indicate cases that commenced in the rst 5 min of
each of the three decision sessions; tick marks on the xaxis denote 5-min intervals save for the last tick mark, which represents all cases that began after more
than 50 min; the dotted lines denote food breaks. We combined all of the cases that began after more than 50 min so that each data point would reect at
least 10 cases.
Table S1. Results of analysis that includes only parole requests
Specication
Variable 1 2 3 4
Overall decision count 0.075*** (0.025) 0.080*** (0.026)
Overall count including nondecisions 0.101*** (0.022) 0.106*** (0.023)
Session 1/decision 1 1.016** (0.435) 0.876** (0.433) ——
Session 1/decision 2 1.444*** (0.451) 1.359*** (0.449) 1.480*** (0.458) 1.389*** (0.448)
Session 1/decision 3 0.007 (0.454) 0.088 (0.452) 0.076 (0.460) 0.117 (0.457)
Session 2/decision 1 1.310*** (0.403) 1.063*** (0.407) 1.288*** (0.407) 1.051** (0.410)
Session 2/decision 2 0.042 (0.411) 0.181 (0.416) 0.031 (0.414) 0.234 (0.417)
Session 2/decision 3 1.035** (0.411) 0.833** (0.415) 0.960** (0.413) 0.780* (0.416)
Session 3/decision 1 3.079*** (0.474) 2.947*** (0.479) 3.001*** (0.475) 2.894*** (0.480)
Session 3/decision 2 1.035** (0.535) 0.859 (0.543) 0.972* (0.538) 0.817 (0.546)
Session 3/decision 3 0.196 (0.800) 0.322 (0.805) 0.286 (0.803) 0.379 (0.807)
Session 1 0.373 (0.309) 0.757** (0.322) 0.526* (0.317) 0.874*** (0.327)
Session 3 1.119*** (0.392) 0.771* (0.403) 0.998** (0.397) 0.681* (0.408)
Severity of offense 0.017 (0.111) 0.030 (0.112) 0.037 (0.117) 0.014 (0.118)
Previous imprisonments 0.193*** (0.063) 0.186*** (0.064) 0.184*** (0.066) 0.176*** (0.066)
Months served 0.003 (0.003) 0.003 (0.003) 0.003 (0.003) 0.003 (0.003)
Rehabilitation program 2.971*** (1.085) 2.944*** (1.098) 2.387** (1.107) 2.372** (1.134)
Ethnicity (0 = Jew, 1 = Arab) 0.182 (0.186) 0.207 (0.187) 0.136 (0.191) 0.158 (0.192)
Sex (0 = male, 1 = female) 0.528 (0.340) 0.532 (0.341) 0.510 (0.345) 0.523 (0.412)
Proportion favorable decisions ——0.704* (0.407) 0.423 (0.411)
2 Log likelihood 794.519 781.166 745.042 732.443
This table is analogous to Table 1, but with a restricted sample size in which only parole requests were included. The results of this xed-effect logistic
regression analysis are very similar to those presented in Table 1. SEs in parentheses. *P<0.10, **P<0.05, ***P<0.01.
Danziger et al. www.pnas.org/cgi/content/short/1018033108 3of10
Table S2. Results of the analysis in which the two judges who most frequently occurred in the
sample are excluded
Specication
Variable 1 2
Overall decision count 0.034 (0.028) 0.043 (0.029)
Session 1/decision 1 0.783* (0.467)
Session 1/decision 2 1.484*** (0.492) 1.563*** (0.496)
Session 1/decision 3 0.789* (0.453) 0.788* (0.456)
Session 2/decision 1 1.071** (0.459) 1.005** (0.458)
Session 2/decision 2 0.339 (0.448) 0.270 (0.449)
Session 2/decision 3 0.739* (0.432) 0.690 (0.433)
Session 3/decision 1 3.000*** (0.526) 2.899*** (0.526)
Session 3/decision 2 0.913* (0.539) 0.807 (0.544)
Session 3/decision 3 0.024 (0.687) 0.083 (0.690)
Session 1 0.018 (0.351) 0.169 (0.357)
Session 3 1.176*** (0.380) 1.012*** (0.386)
Severity of offense 0.249* (0.134) 0.245* (0.137)
Previous imprisonments 0.212*** (0.076) 0.177** (0.076)
Months served 0.008** (0.004) 0.008** (0.003)
Rehabilitation program 2.322*** (0.828) 1.863** (0.877)
Ethnicity (0 = Jew, 1 = Arab) 0.126 (0.199) 0.067 (0.205)
Sex (0 = male, 1 = female) 0.043 (0.349) 0.045 (0.362)
Proportion favorable decisions 0.880** (0.430)
2 Log likelihood 670.646 629.118
Note that this table is based on a smaller sample size (n= 653). We chose to drop these two specic judges and
not others because, due to their relative frequency, they were overrepresented in our sample. SEs in parentheses.
*P<0.10, **P<0.05, ***P<0.01.
Table S3. Results of the analysis presented in which each of the eight judges are excluded one at
a time (excludes Judge 1):
Specication
Variable 1 2
Overall decision count 0.070*** (0.025) 0.080*** (0.026)
Session 1/decision 1 0.636* (0.407)
Session 1/decision 2 1.382*** (0.432) 1.443*** (0.435)
Session 1/decision 3 0.414 (0.390) 0.397 (0.393)
Session 2/decision 1 1.162*** (0.407) 1.144*** (0.407)
Session 2/decision 2 0.288 (0.387) 0.249 (0.387)
Session 2/decision 3 0.835** (0.380) 0.808** (0.380)
Session 3/decision 1 2.465*** (0.455) 2.378*** (0.454)
Session 3/decision 2 0.813* (0.500) 0.732 (0.503)
Session 3/decision 3 0.238 (0.673) 0.317 (0.675)
Session 1 0.186 (0.296) 0.326 (0.300)
Session 3 1.042*** (0.349) 0.904** (0.354)
Severity of offense 0.163 (0.111) 0.161 (0.113)
Previous imprisonments 0.228*** (0.066) 0.185*** (0.065)
Months served 0.005*(0.003) 0.005(0.003)
Rehabilitation program 2.276 (0.813) 1.798** (0.848)
Ethnicity (0 = Jew, 1 = Arab) 0.145 (0.170) 0.136 (0.175)
Sex (0 = male, 1 = female) 0.166 (0.331) 0.098 (0.340)
Proportion favorable decisions 0.843** (0.376)
2 Log likelihood 898.328 845.734
Each table S3S10 presents the results of a xed-effect logistic regression analysis for seven rather than eight
judges. SEs in parentheses. *P<0.10, **P<0.05, ***P<0.01.
Danziger et al. www.pnas.org/cgi/content/short/1018033108 4of10
Table S4. Results of the analysis excluding Judge 2:
Specication
Variable 1 2
Overall decision count 0.059*** (0.022) 0.062*** (0.023)
Session 1/decision 1 1.015** (0.419)
Session 1/decision 2 1.445*** (0.425) 1.492*** (0.430)
Session 1/decision 3 0.665* (0.393) 0.644* (0.397)
Session 2/decision 1 0.941** (0.386) 0.921** (0.390)
Session 2/decision 2 0.274 (0.374) 0.228 (0.378)
Session 2/decision 3 0.653* (0.371) 0.607* (0.374)
Session 3/decision 1 3.410*** (0.493) 3.340*** (0.494)
Session 3/decision 2 0.949** (0.480) 0.867* (0.484)
Session 3/decision 3 0.150 (0.669) 0.237 (0.673)
Session 1 0.287 (0.275) 0.452 (0.284)
Session 3 1.142*** (0.341) 1.013*** (0.347)
Severity of offense 0.080 (0.109) 0.039 (0.115)
Previous imprisonments 0.233*** (0.067) 0.237*** (0.071)
Months served 0.007**(0.003) 0.007**(0.003)
Rehabilitation program 2.572*** (0.830) 2.083** (0.879)
Ethnicity (0 = Jew, 1 = Arab) 0.205 (0.177) 0.136 (0.182)
Sex (0 = male, 1 = female) 0.100 (0.312) 0.030 (0.322)
Proportion favorable decisions 0.979*** (0.368)
2 Log likelihood 906.889 849.150
SEs in parentheses. *P<0.10, **P<0.05, ***P<0.01.
Table S5. Results of the analysis excluding Judge 3:
Specication
Variable 1 2
Overall decision count 0.076*** (0.021) 0.079*** (0.021)
Session 1/decision 1 0.892** (0.389)
Session 1/decision 2 1.250*** (0.392) 1.276*** (0.396)
Session 1/decision 3 0.294 (0.365) 0.252 (0.370)
Session 2/decision 1 0.996*** (0.370) 1.018*** (0.374)
Session 2/decision 2 0.272 (0.349) 0.238 (0.352)
Session 2/decision 3 0.671** (0.351) 0.650* (0.353)
Session 3/decision 1 2.813*** (0.435) 2.735*** (0.435)
Session 3/decision 2 0.937** (0.457) 0.863* (0.460)
Session 3/decision 3 0.330 (0.662) 0.403 (0.664)
Session 1 0.422* (0.254) 0.561** (0.260)
Session 3 1.120*** (0.324) 0.981*** (0.329)
Severity of offense 0.068 (0.098) 0.024 (0.102)
Previous imprisonments 0.236*** (0.061) 0.226*** (0.063)
Months served 0.004 (0.003) 0.003 (0.003)
Rehabilitation program 2.419*** (0.812) 1.897** (0.846)
Ethnicity (0 = Jew, 1 = Arab) 0.209 (0.160) 0.203 (0.165)
Sex (0 = male, 1 = female) 0.044 (0.318) 0.040 (0.323)
Proportion favorable decisions 0.995*** (0.341)
2 Log likelihood 1064.794 1000.208
SEs in parentheses. *P<0.10, **P<0.05, ***P<0.01.
Danziger et al. www.pnas.org/cgi/content/short/1018033108 5of10
Table S6. Results of the analysis excluding Judge 4:
Specication
Variable 1 2
Overall decision count 0.077*** (0.022) 0.077*** (0.022)
Session 1/decision 1 0.704* (0.397)
Session 1/decision 2 1.342*** (0.415) 1.383*** (0.421)
Session 1/decision 3 0.493 (0.385) 0.446 (0.389)
Session 2/decision 1 0.908** (0.380) 0.921** (0.383)
Session 2/decision 2 0.286 (0.364) 0.253 (0.367)
Session 2/decision 3 0.929** (0.370) 0.904** (0.371)
Session 3/decision 1 2.495*** (0.461) 2.424*** (0.461)
Session 3/decision 2 1.088** (0.480) 1.038** (0.482)
Session 3/decision 3 0.179 (0.676) 0.243 (0.679)
Session 1 0.373 (0.261) 0.509* (0.267)
Session 3 1.079*** (0.351) 0.963*** (0.356)
Severity of offense 0.039 (0.100) 0.004 (0.105)
Previous imprisonments 0.260*** (0.065) 0.255*** (0.069)
Months served 0.003 (0.003) 0.003(0.003)
Rehabilitation program 2.105** (0.828) 1.684** (0.862)
Ethnicity (0 = Jew, 1 = Arab) 0.282** (0.165) 0.243 (0.170)
Sex (0 = male, 1 = female) 0.421 (0.341) 0.417 (0.346)
Proportion favorable decisions 1.058*** (0.360)
2 Log likelihood 996.649 936.619
SEs in parentheses. *P<0.10, **P<0.05, ***P<0.01.
Table S7. Results of the analysis excluding Judge 5:
Specication
Variable 1 2
Overall decision count 0.076*** (0.021) 0.077*** (0.021)
Session 1/decision 1 0.701* (0.387)
Session 1/decision 2 1.260*** (0.395) 1.330*** (0.401)
Session 1/decision 3 0.267 (0.362) 0.244 (0.365)
Session 2/decision 1 1.037*** (0.369) 1.051*** (0.372)
Session 2/decision 2 0.210 (0.350) 0.180 (0.352)
Session 2/decision 3 0.796** (0.354) 0.775** (0.355)
Session 3/decision 1 3.021*** (0.456) 2.965*** (0.456)
Session 3/decision 2 0.703 (0.514) 0.638 (0.516)
Session 3/decision 3 0.097 (0.674) 0.152 (0.676)
Session 1 0.286 (0.252) 0.393 (0.257)
Session 3 1.227*** (0.345) 1.124*** (0.350)
Severity of offense 0.009 (0.099) 0.029 (0.103)
Previous imprisonments 0.246*** (0.061) 0.234*** (0.063)
Months served 0.004(0.003) 0.004(0.003)
Rehabilitation program 3.123*** (1.080) 2.600** (1.109)
Ethnicity (0 = Jew, 1 = Arab) 0.219 (0.162) 0.206 (0.167)
Sex (0 = male, 1 = female) 0.192 (0.301) 0.146 (0.307)
Proportion favorable decisions 0.836** (0.345)
2 Log likelihood 1054.537 991.893
SEs in parentheses. *P<0.10, **P<0.05, ***P<0.01.
Danziger et al. www.pnas.org/cgi/content/short/1018033108 6of10
Table S8. Results of the analysis excluding Judge 6:
Specication
Variable 1 2
Overall decision count 0.080*** (0.021) 0.081*** (0.021)
Session 1/decision 1 0.913** (0.388)
Session 1/decision 2 1.541*** (0.407) 1.571*** (0.411)
Session 1/decision 3 0.277 (0.362) 0.221 (0.367)
Session 2/decision 1 1.054*** (0.365) 1.075*** (0.368)
Session 2/decision 2 0.309 (0.346) 0.280 (0.348)
Session 2/decision 3 0.663* (0.346) 0.640* (0.347)
Session 3/decision 1 2.890*** (0.436) 2.807*** (0.436)
Session 3/decision 2 0.913** (0.474) 0.831* (0.475)
Session 3/decision 3 0.654 (0.782) 0.728 (0.783)
Session 1 0.342 (0.253) 0.461* (0.258)
Session 3 1.113*** (0.334) 0.977*** (0.340)
Severity of offense 0.070 (0.101) 0.033 (0.105)
Previous imprisonments 0.240*** (0.060) 0.227*** (0.062)
Months served 0.005*(0.003) 0.005*(0.003)
Rehabilitation program 2.224*** (0.830) 1.895** (0.869)
Ethnicity (0 = Jew, 1 = Arab) 0.178 (0.161) 0.155 (0.166)
Sex (0 = male, 1 = female) 0.142 (0.304) 0.097 (0.310)
Proportion favorable decisions 0.856** (0.352)
2 Log likelihood 1065.101 1003.123
SEs in parentheses. *P<0.10, **P<0.05, ***P<0.01.
Table S9. Results of the analysis excluding Judge 7:
Specication
Variable 1 2
Overall decision count 0.078*** (0.022) 0.079*** (0.022)
Session 1/decision 1 0.973** (0.402)
Session 1/decision 2 1.448*** (0.412) 1.480*** (0.416)
Session 1/decision 3 0.280 (0.375) 0.224 (0.379)
Session 2/decision 1 1.219*** (0.383) 1.234*** (0.386)
Session 2/decision 2 0.258 (0.362) 0.210 (0.365)
Session 2/decision 3 0.837** (0.364) 0.805** (0.366)
Session 3/decision 1 2.890*** (0.467) 2.863*** (0.469)
Session 3/decision 2 0.967** (0.490) 0.913* (0.493)
Session 3/decision 3 1.297 (1.060) 1.371 (1.062)
Session 1 0.355 (0.263) 0.511* (0.269)
Session 3 0.905** (0.354) 0.809** (0.359)
Severity of offense 0.040 (0.102) 0.073 (0.107)
Previous imprisonments 0.225*** (0.063) 0.215*** (0.066)
Months served 0.003(0.003) 0.003 (0.003)
Rehabilitation program 1.791** (0.848) 1.174 (0.872)
Ethnicity (0 = Jew, 1 = Arab) 0.216 (0.170) 0.169 (0.174)
Sex (0 = male, 1 = female) 0.363 (0.334) 0.302 (0.343)
Proportion favorable decisions 0.973*** (0.369)
2 Log likelihood 970.604 910.261
SEs in parentheses. *P<0.10, **P<0.05, ***P<0.01.
Danziger et al. www.pnas.org/cgi/content/short/1018033108 7of10
Table S10. Results of the analysis excluding Judge 8:
Specication
Variable 1 2
Overall decision count 0.113*** (0.024) 0.112*** (0.025)
Session 1/decision 1 1.001** (0.423)
Session 1/decision 2 1.246*** (0.411) 1.283*** (0.417)
Session 1/decision 3 0.356 (0.378) 0.321 (0.382)
Session 2/decision 1 1.151*** (0.390) 1.167*** (0.393)
Session 2/decision 2 0.197 (0.360) 0.169 (0.362)
Session 2/decision 3 0.723** (0.362) 0.705** (0.364)
Session 3/decision 1 3.100*** (0.475) 3.050*** (0.474)
Session 3/decision 2 0.700 (0.501) 0.635 (0.504)
Session 3/decision 3 0.194 (0.676) 0.250 (0.679)
Session 1 0.462* (0.273) 0.596**(0.280)
Session 3 0.829** (0.362) 0.728** (0.367)
Severity of offense 0.041 (0.103) 0.019 (0.107)
Previous imprisonments 0.262*** (0.064) 0.256*** (0.068)
Months served 0.004(0.003) 0.004(0.003)
Rehabilitation program 3.457*** (1.094) 2.940*** (1.139)
Ethnicity (0 = Jew, 1 = Arab) 0.174 (0.169) 0.162 (0.174)
Sex (0 = male, 1 = female) 0.210 (0.330) 0.176 (0.337)
Proportion favorable decisions 0.970*** (0.355)
2 Log likelihood 965.293 909.818
SEs in parentheses. *P<0.10, **P<0.05, ***P<0.01.
Table S11. Nested model comparison tests
Specication
Variable 1 2 3 4
Within session decision count 0.217*** (0.023) 0.201*** (0.024) ——
Severity of offense 0.062 (0.089) 0.029 (0.093) 0.094 (0.085) 0.041 (0.090)
Previous imprisonments 0.250*** (0.056) 0.231*** (0.058) 0.250*** (0.055) 0.231*** (0.057)
Months served 0.002 (0.002) 0.002 (0.002) 0.002 (0.002) 0.001 (0.002)
Rehabilitation program 1.931** (0.773) 1.471* (0.784) 1.681** (0.756) 1.371* (0.774)
Ethnicity (0 = Jew, 1 = Arab) 0.131 (0.146) 0.112 (0.151) 0.053 (0.139) 0.027 (0.145)
Sex (0 = male, 1 = female) 0.443 (0.292) 0.353 (0.301) 0.391 (0.278) 0.306 (0.291)
Proportion favorable decisions 1.400*** (0.304) 1.664*** (0.297)
2 Log likelihood 1239.434 1156.380 1351.338 1241.543
We conducted our xed-effect logistic regression analysis with and without an ordinal position variable and without any of the
session or session/position dummies to ascertain whether adding these variables increased model t using a likelihood ratio test. In all
cases, adding variables that denote ordinal position yield a signicantly better tting model (e.g., compare specications 3 and 4
above with the regressions presented in Table 1; all χ
2
>10, P<0.001). *P<0.10, **P<0.05, ***P<0.01.
Danziger et al. www.pnas.org/cgi/content/short/1018033108 8of10
Table S12. Analysis of linear trend between breaks
Specication
Variable 123456
Within session decision count 0.205*** (0.032) 0.202*** (0.032) ——0.194*** (0.050) 0.193*** (0.050)
Within session decision count
including nondecisions
——0.202*** (0.028) 0.200*** (0.028) ——
Session 1 0.285 (0.291) 0.255 (0.339) 0.191 (0.292) 0.224 (0.340) 0.438 (0.390) 0.642 (0.453)
Session 3 0.749** (0.342) 0.711** (0.343) 0.959*** (0.331) 0.921*** (0.332) 0.536 (0.425) 0.525 (0.427)
Session 1 ×Within session
count
0.022 (0.052) 0.030 (0.057) 0.020 (0.048) 0.036 (0.054) 0.016 (0.074) 0.057 (0.082)
Session 3 ×Within session
count
0.167** (0.080) 0.162** (0.080) 0.101 (0.065) 0.098 (0.066) 0.120 (0.092) 0.111 (0.092)
Severity of offense 0.035 (0.093) 0.008 (0.096) 0.042 (0.094) 0.015 (0.097) 0.248* (0.130) 0.253* (0.133)
Previous imprisonments 0.244*** (0.057) 0.233*** (0.059) 0.244*** (0.057) 0.234*** (0.060) 0.237*** (0.074) 0.207*** (0.073)
Months served 0.002 (0.003) 0.002 (0.003) 0.002 (0.002) 0.002 (0.003) 0.007** (0.003) 0.007* (0.003)
Rehabilitation program 2.114*** (0.791) 1.520* (0.796) 2.052*** (0.797) 1.380* (0.795) 2.048*** (0.801) 1.421* (0.811)
Ethnicity (0 = Jew, 1 = Arab) 0.171 (0.151) 0.146 (0.155) 0.179 (0.153) 0.155 (0.157) 0.151 (0.194) 0.100 (0.199)
Sex (0 = male, 1 = female) 0.122 (0.299) 0.070 (0.306) 0.120 (0.303) 0.056 (0.310) 0.129 (0.346) 0.256 (0.358)
Proportion favorable decisions 0.643** (0.324) 0.554* (0.328) 0.584 (0.416)
2 Log likelihood 1153.638 1090.567 1133.735 1073.548 687.343 647.001
This xed-effect logistic regression analysis tests the robustness of a variable that indicates the ordinal position of a case within a decision session (e.g., after
breakfast snack and until lunch), while controlling for case characteristics. The variables Session 1, Session 3, Rehabilitation Program, Ethnicity, and Sex are
dummy variables as in previous analyses. The negative parameter estimate on the ordinal position variable indicates that the trend(s) apparent in Fig. 1 are
statistically signicant. Specications 5 and 6 drop the two judges with the most observations as in Table S2. SEs in parentheses. *P<0.10, **P<0.05, ***P<0.01.
Table S13. Results of analysis using cumulative minutes elapsed in a session
Specication
Variable 1 2
Cumulative minutes in session 0.021*** (0.005) 0.021*** (0.005)
Session 1 0.223 (0.287) 0.070 (0.294)
Session 3 2.176*** (0.392) 2.008*** (0.395)
Session 1 ×Cumulative minutes 0.002 (0.007) 0.002 (0.007)
Session 3 ×Cumulative minutes 0.015 (0.013) 0.012 (0.013)
Severity of offense 0.014 (0.103) 0.006 (0.103)
Previous imprisonments 0.214*** (0.063) 0.206*** (0.062)
Months served 0.002 (0.003) 0.002 (0.003)
Rehabilitation program 1.920* (1.085) 1.826* (1.087)
Ethnicity (0 = Jew, 1 = Arab) 0.110 (0.166) 0.107 (0.167)
Sex (0 = male, 1 = female) 0.179 (0.321) 0.179 (0.323)
Proportion favorable decisions 1.050*** (0.335)
2 Log likelihood 987.238 976.434
The table presents xed effects logistic regression specications that were conducted to test the effect of
cumulative minutes passed in a decision session on the likelihood of a favorable ruling. The negative and signif-
icant parameters for cumulative minutes suggest that as session times lengthened, judges were more likely to rule
against the prisoner. Note that the second specication controls for the proportion of favorable decisions in the
day (this specication drops the very rst decision of the day). Ethnicity and sex are dummy variables. SEs in
parentheses. *P<0.10, **P<0.05, ***P<0.01.
Danziger et al. www.pnas.org/cgi/content/short/1018033108 9of10
Table S14. Results of analysis using both cumulative minutes and elapsed time in a session
Specication
Variable 1 2
Cumulative minutes in session 0.003 (0.005) 0.002 (0.005)
Within session decision count 0.219*** (0.037) 0.207*** (0.037)
Session 1 0.202 (0.174) 0.117 (0.179)
Session 3 1.810*** (0.233) 1.746*** (0.235)
Severity of offense 0.020 (0.105) 0.015 (0.105)
Previous imprisonments 0.222*** (0.064) 0.215*** (0.063)
Months served 0.003 (0.003) 0.002 (0.003)
Rehabilitation program 1.694 (1.081) 1.660 (1.083)
Ethnicity (0 = Jew, 1 = Arab) 0.108 (0.168) 0.105 (0.169)
Sex (0 = male, 1 = female) 0.011 (0.324) 0.024 (0.325)
Proportion favorable decisions 0.717** (0.342)
2 Log likelihood 948.572 943.428
The table presents xed-effects logistic regression specications that were conducted to test the combined
effect of cumulative minutes elapsed in a decision session and within session decision count on the likelihood of
a favorable ruling. The negative and signicant parameter for decision count, coupled with the nonsignicant
parameter for cumulative minutes, suggests that the critical factor in evoking our order effect is the number of
decisions made rather than the time elapsed. Note that the second specication controls for the proportion of
favorable decisions in the day (this specication drops the very rst decision of the day). Ethnicity and sex are
dummy variables. SEs in parentheses. *P<0.10, **P<0.05, ***P<0.01.
Table S15. Analysis of causal factors in judges decision to take a break
Specication
Variable 1 2
Within session decision count 0.144*** (0.025) 0.152*** (0.027)
Severity of offense 0.053 (0.148) 0.053 (0.149)
Previous imprisonments 0.018 (0.078) 0.013 (0.078)
Months served 0.002 (0.004) 0.003 (0.004)
Rehabilitation program 0.954 (0.649) 1.126 (0.667)
Ethnicity (0 = Jew, 1 = Arab) 0.345 (0.233) 0.341 (0.234)
Sex (0 = male, 1 = female) 0.258 (0.461) 0.164 (0.466)
Proportion favorable decisions 1.284** (0.518)
2 Log likelihood 569.651 558.792
The table presents xed-effects logistic regression specications that were conducted to test determinants of
a judges decision to take a break. None of the variables related to a prisoners case were signicant; that is,
whatever type of case a judge had seen did not prompt his or her desire to take a break. Within session decision
count and a variable that controls for the proportion of favorable decisions in the day (this specication drops the
very rst decision of the day) were signicant. Note that the latter was positive, meaning that as a judge had
made more favorable decisions, he or she was more likely to take a break. Ethnicity and sex are dummy variables.
SEs in parentheses. *P<0.10, **P<0.05, ***P<0.01.
Table S16. Correlations between control variables and ordinal position indicators
Ordinal position variable Severity of offense Previous imprisonments Months served to date Rehabilitation program
Session decision count 0.053 (P= 0.077) 0.027 (P= 0.371) 0.029 (P= 0.340) 0.028 (P= 0.346)
Session count including nondecisions 0.033 (P= 0.274) 0.035 (P= 0.242) 0.010 (P= 0.734) 0.015 (P= 0.615)
Cumulative minutes in session 0.035 (P= 0.280) 0.013 (P= 0.682) 0.004 (P= 0.905) 0.022 (P= 0.491)
Overall decision count 0.081 (P= 0.007) 0.062 (P= 0.038) 0.047 (P= 0.115) 0.017 (P= 0.570)
Overall count including nondecisions 0.047 (P= 0.117) 0.075 (P= 0.012) 0.012 (P= 0.692) 0.011 (P= 0.719)
Pearson correlation coefcients between the ordinal position variables and the control variables used in our regressions (Pvalues appear in parentheses).
Columns refer to the different control variables used in our subsequent regression analyses. Rows refer to different representations of ordinal position.
Danziger et al. www.pnas.org/cgi/content/short/1018033108 10 of 10
... Hunger and rest breaks -effects on rationality -The research conducted by Danziger, Levav and Avnaim-pesso (2011) indicated that some judicial deliberations are not completely rational, there are odd variables that should not have an influence on legal decisions and would condition the decision making of experient judges. In the study, it was found that the percentage of favorable decisions has fallen gradually from ± 65% to almost zero, returning abruptly to ± 65% after a break. ...
... In the study, it was found that the percentage of favorable decisions has fallen gradually from ± 65% to almost zero, returning abruptly to ± 65% after a break. There is evidence that mental function can be restored, and mental fatigue overcome, in part, by interventions such as viewing nature scenes, taking a brief rest, experimenting with positive mood, and increasing body glucose levels (food intake) (Danziger et al., 2011). ...
... These results are in line with the studies of Malik et al. (2008), in which the effects of the Ghrelin hormone over the hunger feeling, as well as the mood swings proportional to this feeling and tendency to greater irritability to emerge as justification of some interviewees' agreement, these factors are also highlighted in the research of Danziger, Levav e Avnaim-pesso (2011). One of the interviewees identifies a direct reference to studies that link the hunger feeling and its influence on decisions, citing the research by Danziger, Levav and Avnaim-pesso (2011). It was found that the majority of the interviewees considered as true the affirmation that eating affects the decisions in the workplace, some of them even brought references to knowledge about physical, biological, and emotional factors triggered in the individuals. ...
Conference Paper
Full-text available
This paper analyzes the perception of leaders about the discoveries of organizational cognitive neuroscience (OCN) and their practical applications in the work environment, as well as identifying gaps and distortions of knowledge that leaders have about this area of study. Based on a study with the participation of one hundred and thirty-six respondents, of which part consider themselves in leadership positions, conducted in two stages, using a quantitative and qualitative method, it was possible to highlight the perception of such leaders about the practices currently used in their companies, even though they are unaware of the neurological factors and reflexes that trigger them. The results also made it possible to propose initiatives and actions for leaders regarding the expansion of knowledge about NCO. Finally, this research presents recommendations for including or changing internal and external leadership training programs on the topics covered in Organizational Cognitive Neuroscience. ABSTRACT This paper analyzes the perception of leaders about the discoveries of organizational cognitive neuroscience (OCN) and their practical applications in the work environment, as well as identifying gaps and distortions of knowledge that leaders have about this area of study. Based on a study with the participation of one hundred and thirty-six respondents, of which part consider themselves in leadership positions, conducted in two stages, using a quantitative and qualitative method, it was possible to highlight the perception of such leaders about the practices currently used in their companies, even though they are unaware of the neurological factors and reflexes that trigger them. The results also made it possible to propose initiatives and actions for leaders regarding the expansion of knowledge about NCO. Finally, this research presents recommendations for including or changing internal and external leadership training programs on the topics covered in Organizational Cognitive Neuroscience.
... Bon, 2006). Plus récemment, des études suggèrent que les juges au tribunal rendraient des jugements biaisés selon s'ils viennent d'avoir une pause repas ou non (Danziger et al., 2011(Danziger et al., , 2015. En laboratoire, notre équipe a montré que le fait d'écouter une musique jugée agréable en même temps que l'on doit juger le plaisir procuré par l'observation d'un tableau biaisait la perception du plaisir esthétique associé au tableau en fonction du jugement porté sur la musique entendue . ...
Thesis
Tous les jours, nous prenons des décisions sur les actions que nous souhaitons entreprendre. Ces décisions se fondent sur un compromis entre les bénéfices que nous espérons obtenir après avoir effectué ces actions, et les coûts, en termes d’effort, associés à ces actions. Cette thèse s’intéresse aux bases cérébrales du compromis coûts/bénéfices au travers de trois études menées chez des participants sains à l’aide de l’imagerie par résonance magnétique fonctionnelle. Dans la première étude, nous avons pu dissocier les bases cérébrales du calcul du compromis coûts/bénéfices des bases cérébrales des variables régulant ce calcul. En effet, dans cette étude, le calcul du compromis coûts/bénéfices était associé au cortex préfrontal ventromédian alors que la confiance dans la décision et le temps passé à délibérer étaient associés à des parties plus dorsales du cortex préfrontal médian. La seconde étude a permis de montrer que, dans deux tâches, impliquant un effort mental ou physique, la performance s’expliquait mieux par un biais pavlovien, donnant plus de poids aux gains qu’aux pertes, que par une aversion à la perte, telle qu’elle a été caractérisée principalement dans des tâches de choix. La troisième étude nous a permis de montrer que, même dans une tâche simple d’apprentissage par renforcement, les aires cérébrales liées à l’exécution d’un effort mental étaient recrutées au moment du calcul du compromis coûts/bénéfices, suggérant que cette tâche n’était pas effectuée de manière purement automatique. L’ensemble de nos résultats permet de mieux caractériser les aires cérébrales impliquées dans le compromis coûts/bénéfices et les conditions dans lesquelles ces aires sont actives.
... Counter-intuitive effects of bodily states, such as hunger and thirst, on decision-making have been extensively studied [152]. A salient example of the latter can be seen in the context of legal judgements [153], where the homeostatic status of judges has been found to influence the severity of their judgements; all other things being equal, a person does not want to receive their verdict just before lunch. Similarly, if a person tries to switch off the signal without addressing the cause, perhaps by taking a drug to modify pain, they are likely to experience a recurrence of the signal when the drug wears off, which is one reason why pain pills can become so addictive. ...
Article
Full-text available
Organisations are aware of the need to maintain the mental health of their employees. People’s capacity to recognise and manage their moods and emotions is critical to sustainable mental health, performance, and quality of life, while failure to do so can result in underperformance, disengagement, and in some cases, mental illness. Employees of organisations that provide an appropriate strategy and support are likely to experience sustained psychological and mental health benefits. In this paper, we synthesise previous research into a theoretical framework distinguishing mood from emotion via both top-down (cognitive) and bottom-up (biological) factors. We propose a 4R model to help individuals Recognise a mood as distinct from an emotion, or vice-versa, and respond in one of three ways; Restore, Resolve, or Regulate. The model posits mood as an interoceptive signal of internal biological homeostasis, and emotion as a signal of external, often social, events that disrupt homeostasis; mood and emotion serve as internal and external bio-affective feedback loops, respectively. We propose that mood is modified positively by the restoration of homeostasis, whereas emotion is modified positively by behavioural resolution of the emotion-eliciting event. The 4R model is low-cost, preventative, and can be applied peer-to-peer in organisations without expert supervision.
... Surely, a positive bias towards automation could be defended in all these cases. If we think about how wrong human inferences can be (Kahneman 2011) or about how a judges' eating habits may affect the harshness of punishment (Glöckner 2016;Danziger et al. 2011), the idea of automating the process may quickly arise as the fairest as well as the most efficient prospect available. However, a judgement made through an algorithm is not always better than a judgment performed by a human being, as we already explained above, especially because-in the end-humans are strongly biased creatures, who invariably prefer evaluations made by other humans (especially when it comes to punishment) (McKay 2020). ...
Article
Full-text available
Advances in artificial intelligence and robotics stand to change many aspects of our lives, including our values. If trends continue as expected, many industries will undergo automation in the near future, calling into question whether we can still value the sense of identity and security our occupations once provided us with. Likewise, the advent of social robots driven by AI, appears to be shifting the meaning of numerous, long-standing values associated with interpersonal relationships, like friendship. Furthermore, powerful actors’ and institutions’ increasing reliance on AI to make decisions that may affect how people live their lives may have a significant impact on privacy while also raising issues about algorithmic transparency and human control. In this paper, building and expanding on previous works, we will look at how the deployment of Artificial Intelligence technology may lead to changes in identity, security, and other crucial values (such as friendship, fairness, and privacy). We will discuss what challenges we may face in the process, while critically reflecting on whether such changes may be desirable. Finally, drawing on a series of considerations underlying virtue ethics, we will formulate a set of preliminary suggestions, which—we hope—can be used to more carefully guide the future roll out of AI technologies for human flourishing; that is, for social and moral good.
... Judges respond markedly to workplace disruptions and other extraneous factors when deciding on case dispositions and sentencing outcomes (Philippe and Ouss, 2018;Eren and Mocan, 2018;Danziger et al., 2011;Shumway and Wilson, 2022;Englich and Soder, 2009;Simon, 2012;Dijksterhuis et al., 1996;Wyer and Carlston, 2018;Guthrie et al., 2007). 4 For instance, Eren and Mocan (2018) find that an unforeseen loss in a local football game increases juvenile court sentences handed down by judges by nearly 6.5 percent. ...
Preprint
Full-text available
High temperatures have been shown to affect human cognition and decision-making in a variety of settings. In this paper, we explore the extent to which higher temperatures affect judicial decision-making in India. We use data on over 3 million judicial outcomes from the Indian eCourt platform, merged with high-resolution gridded daily weather data. We estimate causal effects leveraging a fixed effects framework. We find that high daily maximum temperatures raise the likelihood of convictions. Our results are robust to numerous controls and specification changes. Our findings contribute to a growing literature that documents that the negative impacts of rising temperatures are often more severe in low-and middle-income countries. JEL codes: K37, K41, Q54.
... Studies have revealed implicit biases in the courtroom from previously unexamined factors, for example, differences in how severely a judge sentences a defendant on the basis of whether the sentencing occurs before or after the judge has had lunch. 357 Current strategies to mitigate implicit bias in the courtroom include acknowledging the existence of implicit bias, routinely examining thought processes and decisions to check for possible bias, and increasing exposure to stereotyped group members. 358 Jurors are unlikely to recognize or reveal their own implicit biases. ...
... In addition, young people who participate in youth programs can gain knowledge and skills in areas such as self-expression, exploration of interests, social justice, critical consciousness, and development and appreciation of social identities including racial identity (Carey et al., 2020;Halpern, 2003;Ladson-Billings, 2014;The National Child Traumatic Stress Network, 2017). health, and commitment (Collie et al., 2012;Danziger et al., 2011;Herman et al., 2018;Jennings & Greenberg, 2009;Mosely, 2018;Osher et al., 2020;Sandilos et al., 2018). ...
Article
Full-text available
Background Youth-serving organizations in the United States provide programs, activities, and opportunities for young people before school, during school, after school, in summer, and on weekends. At the core of youth-serving organizations are the adults; that is, youth development staff.Objective In this explanatory sequential mixed methods study we explored youth development staff’s stress and worries, their compassion satisfaction, and whether stress and compassion satisfaction varied by race/ethnicity and gender during the early months of the COVID-19 pandemic – a collective trauma event.Methods We surveyed 283 youth development staff and interviewed a subset of 25.ResultsResults suggest that youth development staff experienced stress and compassion satisfaction during the COVID-19 pandemic.Conclusion We recommend organizational leaders provide youth development staff with support before a collective trauma event. They can work to change, add, or remove policies, practices, and routines to help decrease stress and increase compassion satisfaction. In addition, based on our results from this study our primary recommendation specific to collective trauma events, after taking care of their own personal wellness, is for youth development staff to focus on what is in their control and work to do those things for as many young people as they can.
Article
A central question in development economics is whether there are adequate checks and balances on the executive. This paper provides causal evidence of how increasing constraints on the executive—via removal of presidential discretion in judicial appointments—promotes the rule of law. The age structure of judges at the time of the reform and the mandatory retirement age law provide us with an exogenous source of variation in the termination of presidential discretion in judicial appointments. Overall, the results indicate that presidential appointment of judges deteriorates the rule of law. Even one degree of separation between the judiciary and the president matters. (JEL D72, K40, O17)
Article
We add to our understanding of the optimal method of judicial selection by exploiting an unusual feature in North Carolina: judges rotate location every 6 months. This allows us to identify the existence and source of sentencing variation over the electoral cycle. We show that when elections approach, felony sentences rise. This increase is found exclusively when judges are sentencing in their district of election, and only when elections are contested. When judges hear cases outside their home district, sentences do not significantly vary over the electoral cycle. Our results show that electoral sentencing cycles can be explained by strategic sentencing by judges in an attempt to please voters. The unique setting allows us to reject alternative behavioral or contextual explanations for the rise in sentences as elections approach. (JEL K42).
Article
Making choices, responding actively instead of passively, restraining impulses, and other acts of self-control and volition all draw on a common resource that is limited and renewable, akin to strength or energy. After an act of choice or self-control, the self's resources have been expended, producing the condition of ego depletion. In this state, the self is less able to function effectively, such as by regulating itself or exerting volition. Effects of ego depletion appear to reflect an effort to conserve remain ing resources rather than full exhaustion, although in principle full exhaustion is possible. This versatile but limited resource is crucial to the self's optimal functioning, and the pervasive need to conserve it may result in the commonly heavy reliance on habit, routine, and automatic processes.
Article
Two experiments investigated how people replenish the self's limited regulatory resource after it is depleted by self-control exertion. Specifically, in Experiment 1, when depleted participants received a 10-minute period between regulatory tasks, their subsequent performance equaled non-depleted participants. In Experiment 2, inducing participants to relax between self-regulation tasks reduced the typical depletion effects. Thus, these findings suggest that replenishment of the self's depleted resources occurs given the occurrence of favorable conditions.
Article
How do judges judge? Do they apply law to facts in a mechanical and deliberative way, as the formalists suggest they do, or do they rely on hunches and gut feelings, as the realists maintain? Debate has raged for decades, but researchers have offered little hard evidence in support of either model. Relying on empirical studies of judicial reasoning and decision making, we propose an entirely new model of judging that provides a more accurate explanation of judicial behavior. Our model accounts for the tendency of the human brain to make automatic, snap judgments, which are surprisingly accurate, but which can also lead to erroneous decisions. Equipped with a better understanding of judging, we then propose several reforms that should lead to more just and accurate outcomes.