diff --git a/R/compute_indirect_effect_for.R b/R/compute_indirect_effect_for.R
index bffb750..e991f79 100644
--- a/R/compute_indirect_effect_for.R
+++ b/R/compute_indirect_effect_for.R
@@ -29,11 +29,11 @@
 #'   coefficients in a linear regression, \eqn{a \times b}{a * b} is now the
 #'   indirect effect we wanted to compute (see the Models section).
 #'
-#'   Thanks to the returned values of \eqn{a}{a} and {b}{b} (\eqn{b_51}{b_51}
-#'   and \eqn{b_64}{b_64}, see the Models section), it is now easy to compute
-#'   \eqn{a \times b}{a * b}. `compute_indirect_effect_for` uses the same
-#'   approach than the [`add_index`] function. A Monte Carlo simulation is used
-#'   to compute the indirect effect index (MacKinnon et al., 2004).
+#'   Thanks to the returned values of \eqn{a}{a} and \eqn{b}{b}
+#'   (\eqn{b_51}{b_51} and \eqn{b_64}{b_64}, see the Models section), it is now
+#'   easy to compute \eqn{a \times b}{a * b}. `compute_indirect_effect_for` uses
+#'   the same approach than the [`add_index`] function. A Monte Carlo simulation
+#'   is used to compute the indirect effect index (MacKinnon et al., 2004).
 #'
 #' @section Models: In a moderated mediation model, three models are used.
 #'   `compute_indirect_effect_for` uses the same model specification as
diff --git a/R/mdt_simple.R b/R/mdt_simple.R
index faf1145..f51ee5a 100644
--- a/R/mdt_simple.R
+++ b/R/mdt_simple.R
@@ -28,7 +28,7 @@
 #'
 #'   with \eqn{c} the total effect of \eqn{X} on \eqn{Y}, \eqn{c'} the direct of
 #'   \eqn{X} on \eqn{Y}, and \eqn{ab} the indirect effect of \eqn{X} on \eqn{Y}
-#'   through {M} (see Models section).
+#'   through \eqn{M} (see Models section).
 #'
 #'   To assess whether the indirect effect is different from the null, one has
 #'   to assess the significance against the null for both \eqn{a} (the effect of
diff --git a/man-roxygen/within_details.R b/man-roxygen/within_details.R
index 54e1274..88fec2c 100644
--- a/man-roxygen/within_details.R
+++ b/man-roxygen/within_details.R
@@ -15,7 +15,7 @@
 #'
 #'   with \eqn{c} the total effect of \eqn{X} on \eqn{Y}, \eqn{c'} the direct of
 #'   \eqn{X} on \eqn{Y}, and \eqn{ab} the indirect effect of \eqn{X} on \eqn{Y}
-#'   through {M} (see Models section).
+#'   through \eqn{M} (see Models section).
 #'
 #'   To assess whether the indirect effect is different from the null, one has
 #'   to assess the significance against the null for both \eqn{a} (the effect of
diff --git a/man/compute_indirect_effect_for.Rd b/man/compute_indirect_effect_for.Rd
index 49b1d86..2440d29 100644
--- a/man/compute_indirect_effect_for.Rd
+++ b/man/compute_indirect_effect_for.Rd
@@ -44,11 +44,11 @@ set. Because of the new coding, and because of how one interprets
 coefficients in a linear regression, \eqn{a \times b}{a * b} is now the
 indirect effect we wanted to compute (see the Models section).
 
-Thanks to the returned values of \eqn{a}{a} and {b}{b} (\eqn{b_51}{b_51}
-and \eqn{b_64}{b_64}, see the Models section), it is now easy to compute
-\eqn{a \times b}{a * b}. \code{compute_indirect_effect_for} uses the same
-approach than the \code{\link{add_index}} function. A Monte Carlo simulation is used
-to compute the indirect effect index (MacKinnon et al., 2004).
+Thanks to the returned values of \eqn{a}{a} and \eqn{b}{b}
+(\eqn{b_51}{b_51} and \eqn{b_64}{b_64}, see the Models section), it is now
+easy to compute \eqn{a \times b}{a * b}. \code{compute_indirect_effect_for} uses
+the same approach than the \code{\link{add_index}} function. A Monte Carlo simulation
+is used to compute the indirect effect index (MacKinnon et al., 2004).
 }
 \section{Models}{
  In a moderated mediation model, three models are used.
diff --git a/man/mdt_simple.Rd b/man/mdt_simple.Rd
index f4e8776..572661d 100644
--- a/man/mdt_simple.Rd
+++ b/man/mdt_simple.Rd
@@ -61,7 +61,7 @@ The total effect of \eqn{X} on \eqn{Y} can be described as follows:
 
 with \eqn{c} the total effect of \eqn{X} on \eqn{Y}, \eqn{c'} the direct of
 \eqn{X} on \eqn{Y}, and \eqn{ab} the indirect effect of \eqn{X} on \eqn{Y}
-through {M} (see Models section).
+through \eqn{M} (see Models section).
 
 To assess whether the indirect effect is different from the null, one has
 to assess the significance against the null for both \eqn{a} (the effect of
diff --git a/man/mdt_within.Rd b/man/mdt_within.Rd
index 4c0ee79..0d8d64d 100644
--- a/man/mdt_within.Rd
+++ b/man/mdt_within.Rd
@@ -73,7 +73,7 @@ conceptually described as follows:
 
 with \eqn{c} the total effect of \eqn{X} on \eqn{Y}, \eqn{c'} the direct of
 \eqn{X} on \eqn{Y}, and \eqn{ab} the indirect effect of \eqn{X} on \eqn{Y}
-through {M} (see Models section).
+through \eqn{M} (see Models section).
 
 To assess whether the indirect effect is different from the null, one has
 to assess the significance against the null for both \eqn{a} (the effect of
diff --git a/man/mdt_within_wide.Rd b/man/mdt_within_wide.Rd
index abcca44..7fcf2fc 100644
--- a/man/mdt_within_wide.Rd
+++ b/man/mdt_within_wide.Rd
@@ -70,7 +70,7 @@ conceptually described as follows:
 
 with \eqn{c} the total effect of \eqn{X} on \eqn{Y}, \eqn{c'} the direct of
 \eqn{X} on \eqn{Y}, and \eqn{ab} the indirect effect of \eqn{X} on \eqn{Y}
-through {M} (see Models section).
+through \eqn{M} (see Models section).
 
 To assess whether the indirect effect is different from the null, one has
 to assess the significance against the null for both \eqn{a} (the effect of